US government officials have revealed that Chinese state-sponsored hackers breached a host of federal institutions as part of a recent espionage campaign.

In a statement, the US Justice Department and FBI confirmed it has successfully seized domains used to support two hacking platforms known as ‘QScan’ and ‘QTRouter’.

These platforms were used by a state-sponsored group known as ‘QTFY’ to target critical infrastructure and “other sensitive networks”.

According to the FBI, the group successfully breached NASA, the Federal Reserve, Department of Energy, the US Senate, Department of Justice, and other agencies.

The scope and nature of the breaches has not been revealed.

“Federal law enforcement investigated and disabled the PRC’s malicious software, the latest in a series of technical operations to dismantle indiscriminate hacking activities sponsored by the People’s Republic of China,” said Attorney General Todd Blanche.

“State-sponsored malicious hackers preying on America’s critical infrastructure will be stopped and prosecuted. We are here to ensure security for the American people and will use every tool we have to keep that promise.”

State-backed operators

The QTFY group provides hacker-for-hire services to paying customers, as well as the People’s Republic of China’s (PRC) Ministry of State Security and the People’s Liberation Army (PLA), according to court documents.

QScan and QTRouter are frequently-used platforms in QTFY operations, according to the Justice Department. The first of these is used to scan for and automatically infect vulnerable IoT devices, which are then added to the broader QTRouter network of controlled devices.

“QTRouter consists of these compromised IoT devices, as well as commercial proxy service devices and leased virtual private servers,” the Justice Department noted.

QTRouter serves as an "obfuscation network" for the group, meaning that it enables QTFY and other "malicious cyber actors" to conceal their location and activities.

Officials noted that the seized domains were hard-coded into both QScan and QTRouter malware, which were then used in follow-up attacks against US government agencies.

Disrupting cyber threats

The takedown announced by the Justice Department marks the latest in a string of operations aimed at mitigating Chinese state-backed threats.

In 2025, for example, the FBI helped remove the PlugX surveillance malware from more than 4,000 computers across the US following a campaign conducted by the Mustang Panda group.

That operation came just months after a similar sting that crippled a botnet hosted by the Flax Typhoon hacker group.

“These tools were used by PRC cyber actors to hide the origin of their attacks,” said FBI Director Kash Patel.

“Thanks to the work of FBI San Diego, FBI Cyber Division, and DOJ partners, we seized adversary infrastructure and shut these platforms down.”

FOLLOW US ON SOCIAL MEDIA

Google has unveiled a series of new FinOps features for Gemini Enterprise as the tech giant aims to help developers manage surging AI costs.

Users of the AI service will be offered “expanded billing flexibility” and new cost management tools for agents, the company revealed.

This includes flexible payment options that allow “predictable per-user seat subscriptions” on a pay-as-you-go basis. Google said this will allow users to run agent workloads without hitting token quota limits mid-task.

A flexible savings plan option will also allow enterprises to set variable monthly spend limits to accommodate for growing use rates. This, the company claimed, will help reduce token consumption rates and overall costs.

“If your organization has steady or growing AI workloads, Gemini Enterprise Flexible Savings Plans offer a simple, spend-based commitment model across Gemini Enterprise usage,” the company said in an announcement.

“FSPs are designed to lower token costs while keeping budgets flexible.”

Elsewhere, new spending guardrails will allow teams to set “hard monthly caps” on AI spend on a project-by-project basis. These features allow users to estimate agent costs and prevent sudden budget spikes, the company said.

Tackling AI costs

The move by Google comes amidst growing concerns over rising AI costs in recent months.

As ITPro reported earlier this month, Gartner projects surging agentic AI adoption rates to have a significant impact on enterprise infrastructure, with inference spending in particular expected to increase dramatically.

A key factor behind these rising costs lies in how agents operate compared to more traditional chatbot-based interactions. Agents essentially operate in loops, rather than through a single request and response-type approach with chatbots – and that requires far more processing power.

Research from Signal65, for example, found that agentic AI workloads consume anywhere between four-to-fifteen times more tokens compared to chatbots.

In July, Patrick Brogan, director of the FinOps advisory team at DevOps firm Harness told ITPro that rising AI costs mean FinOps practices are more important than ever and could help keep a lid on bills.

Developer spending features

Google has also sharpened its focus on developer spending rates with the integration of its Antigravity platform within Gemini Enterprise.

This will allow developers to leverage established quotas within an enterprise subscription, creating closer alignment in terms of budgeting and consumption rates.

“To be more efficient with agentic coding costs, we are pooling developer tools quota included in each Gemini Enterprise subscription and making it available across the whole Google Cloud project so your teams can benefit from the capacity you’re already purchasing,” Google said in a statement.

Tighter budget and cost controls come at a crucial time for developers, with the use of agents in this domain accelerating rapidly over the last two years. Research from Gartner projects that AI costs could exceed the average developer salary by 2028, as ITPro reported in July.

Speaking to ITPro at the time, Gartner senior principal analyst Nitish Tyagi said a stronger focus on cost optimization and budget controls will be crucial to avoid overspending by AI-savvy developer teams.

FOLLOW US ON SOCIAL MEDIA

Hundreds of internet-facing Zimbra instances have been compromised through a vulnerability that was patched last month, researchers have warned.

The Zimbra Collaboration Suite (ZCS) hosts email, calendars, contacts, and administrative services. It has hundreds of millions of users, including thousands of businesses and hundreds of government agencies worldwide.

The security flaw, tracked as CVE-2026-73570 and rated high severity with a CVSS score of 8.9, allows unauthenticated attackers to execute malicious code remotely.

It works by exploiting a command injection weakness in the SNMP monitoring component when SNMP notifications are enabled, and affects Zimbra Collaboration's Simple Network Management Protocol (SNMP).

This is a monitoring functionality in deployments where the optional zimbra-snmp package is installed and SNMP notifications are enabled.

"This vulnerability could allow an unauthenticated attacker to send specially crafted SMTP requests, potentially resulting in the execution of arbitrary operating system commands as the Zimbra user when the optional zimbra-snmp package is installed and SNMP notifications are enabled," warned the Hong Hong Computer Emergency Response Team (HKCert) this week.

Hackers can then establish persistence, access email accounts, harvest credentials, and move laterally to other systems. Synacor, the company behind the collaboration suite, was first made aware of the issue in June, issued a temporary mitigation, and fixed it with the release of ZCS version 10.1.20 on July 20.

Agencies issue Zimbra flaw warnings

The US Cybersecurity and Infrastructure Security Agency (CISA) last week added the flaw to its Known Exploited Vulnerabilities (KEV) list and ordered US federal civilian agencies to address it within three days.

However, it appears that the vulnerability is still being actively exploited.

Non-profit security organization Shadowserver said it has spotted at least 274 internet-exposed Zimbra instances that have been breached.

Meanwhile, at least 8,200 organizations worldwide are still using vulnerable versions, although this doesn't mean they're exploitable as the vulnerability may be in a non-default configuration.

Dray Agha, senior manager of Huntress’ EMEA security operations center, said the exploitation of the flaw highlights the importance of rapid patching.

"This widespread compromise of Zimbra servers is a textbook example of the enterprise patching gap. The patch for CVE-2026-73570 was released in July, yet weeks later, attackers are still easily finding hundreds of vulnerable instances to exploit," he said.

"When dealing with an unauthenticated, remote code execution flaw on an internet-facing email server, the window for remediation isn't measured in weeks or days, it’s honestly measured in hours. Organizations need to treat collaboration suites as highly critical perimeter infrastructure and patch them with zero delay."

It's not clear who is behind the attacks, but exploitation of Zimbra vulnerabilities has in the past been linked to Russian state-sponsored hackers, including APT28, APT29, and Winter Vivern.

These groups have targeted military and diplomatic intelligence, alongside opportunistic cybercriminals seeking financial gain.

FOLLOW US ON SOCIAL MEDIA

New research from McKinsey suggests enterprises are finally reporting benefits from AI, but cost-related concerns and “constrained” productivity gains still weigh heavy on the minds of IT leaders.

Figures from the consultancy’s State of AI in 2026 report show more than one-third (37%) of respondents recorded “some” earnings impact with the technology, although that marks the same share from last year’s report.

Meanwhile, “AI high performers” said AI is now having a “significant” impact on operational efficiency and earnings.

The report suggests that enterprise AI maturity is rising, with an increasing number of enterprises now successfully scaling the technology. Improvements in this regard are being reported across the board, McKinsey noted, with respondents scaling chatbots, coding agents, and more sophisticated AI agents.

The use of agentic AI has increased significantly, for example, with 40% of large enterprises scaling agents, up from 27% last year. Meanwhile, around two-in-ten are actively scaling software coding agents.

“Among AI tools, chatbots are the most widely scaled, with 47 percent of respondents saying their organizations are scaling them across the enterprise,” the report states.

“About two-in-ten respondents report reaching the scaling phase across their organization with AI agents and a similar share report the same with software coding agents.”

Mismatched AI productivity

Workforce productivity gains rank among the most notable improvements for enterprises, according to McKinsey.

Around 80% of respondents said that AI has improved individual productivity, while 50% report that AI helps them “make better decisions”. Crucially, however, McKinsey warned that “the experience is not universally positive”.

Mid-level managers and individual workers are more likely than executives to report AI-related problems. McKinsey said this shows that AI impact “remains concentrated” among certain groups within the enterprise.

This is by no means the first study to highlight this trend. Research from Accenture in April this year found many enterprises are dealing with mismatched AI productivity gains.

Similar to McKinsey, the report found that workers are reporting benefits with the technology such as higher-quality output levels and faster project delivery times.

Yet the report pointed to a “widening gap” between the basic use of the technology and its wider enterprise impact. A key factor, Accenture said, lies in the fact that organizations' process changes are “lagging behind” integration rates.

Similarly, many AI strategies typically fail to modernize underlying IT infrastructure or workflows to accommodate the technology.

“A striking finding this year is the gap between individual gains and enterprise impact. At the individual level, AI is clearly a boon: 80 percent of survey respondents say it has improved their productivity and half say it helps them make better decisions,” said Dan Tinkoff, senior partner at McKinsey.

“Yet, only 37 percent of organizations report any positive EBIT contribution, essentially flat compared with last year.”

AI costs are taking the shine off enterprise gains

Cost-related concerns continue to weigh heavy on IT leaders’ minds, according to McKinsey. Around 20% of respondents said that AI-related operating costs have “constrained their AI use”.

“Those cost constraints are being reported across the full range of AI tools. For each of three tools – AI chatbots, AI agents, and software coding agents – about one-in-ten respondents say their organizations’ use has been constrained by costs,” the report states.

McKinsey also highlighted token-related cost concerns. This has become a recurring pain point for some businesses over the last year as firms ramp up adoption of the technology.

The ‘tokenmaxxing’ trend, for example, has prompted a surge in usage and landed some firms with hefty bills.

As ITPro previously reported, Accenture asked workers to stop using AI for basic tasks amidst skyrocketing costs while Uber revealed it blew through its entire annual AI budget in just four months.

Rising operational costs haven’t dampened investment rates, however. More than one-quarter (28%) of respondents said their organisation is now spending more than 10% of their total IT budget on AI tools.

“Looking ahead, 60 percent of respondents expect their organizations to increase their AI investments over the next year,” McKinsey noted.

“Respondents in pharmaceuticals and medical products, insurance, and banking and other financial institutions are the most likely to expect increasing investment.”

FOLLOW US ON SOCIAL MEDIA

Most business and IT leaders believe that sovereign AI is important, but they're not so clear on what it actually is and what it means for their business.

IDC defines sovereign AI as the ability for an organization to have “free choice and control over the design, development, deployment, accessibility, operation, maintenance, and governance of its AI systems and applications, as well as the underlying technology foundations they depend on”.

However, in a survey conducted by IDC on behalf of Cohere, researchers found that one-in-three had difficulty describing sovereign AI in their own words.

Of those who could, 52% of leaders explicitly described it in terms of local or national control, with 35% invoking digital independence.

Individual roles made a difference here, with line-of-business (LOB) leaders primarily viewing sovereign AI as a means for managing business risk, including data security, privacy, and cost controls.

IT leaders, by contrast, focused on regulatory compliance, ensuring that systems meet national and regional requirements. IT professionals were twice as aware as LOB leaders.

IDC predicts that by 2028, CIOs at multinational organizations will boost investments in modular, sovereign-ready cloud and data localization environments by 65% in a bid to future-proof operations against rising sovereignty demands.

Yet adoption remains uneven, according to IDC. The study pointed to a lack of a clear blueprint for operationalizing sovereignty, a fragmented and inconsistent grasp of the concept, and a lack of strategic vision.

On the right track with sovereign AI

Even without an agreed-upon definition, researchers found a clear consensus on the importance of privacy and security.

"Across every industry surveyed, enterprise leaders identified data leakage, privacy, compliance, and regulatory risk as the top concerns that sovereign AI initiatives can address, they said. "Furthermore, the largest enterprises with $20 billion in revenue are much more risk-aware than small firms."

Many see sovereign AI as presenting a competitive advantage - 35% in Canada, 28% in the US, 23% in Germany and 18% in the UK. This is most evident within the telco sector, at 37%, followed by manufacturing at 32%, healthcare at 28%, and financial services and energy, both at 21%.

“Across regulated industries and geopolitically sensitive markets, enterprises are re-architecting their digital operating models — shifting from a global-by-default posture to a deliberately sovereign-by-design approach,” researchers said.

Defining sovereign AI

Cohere said enterprises and governments should define sovereign AI in operational terms, quantify its business value, establish ownership across leadership teams, and build systems that remain available and secure regardless of external events.

Late last year, a global survey from McKinsey found that 71% of executives, investors, and government officials characterized sovereign AI as an 'existential concern' or 'strategic imperative' to their organizational goals.

As a result, the firm projected that sovereign AI could become a $600 billion market by 2030, driven largely by investment in the public sector and regulated industries.

FOLLOW US ON SOCIAL MEDIA

Apple has unveiled new Mac Mini and Mac Studio models, boasting an “extraordinary leap in performance and AI capabilities”.

The new devices wil be equipped with Apple’s M6 and M5 Ultra chips, the latter of which is described as its most powerful to date.

“Today we’re debuting the next giant leap in performance and AI compute for Apple silicon with the incredibly advanced M6 and the most powerful M-series chip yet,” M5 Ultra,” said Sri Santhanam, Apple’s vice president of Silicon Engineering Group.

The move by Apple comes amidst a sharpened focus on local, on-device AI capabilities for developers globally. The Mac Mini, for example, has emerged as a go-to option on this front, enabling users to run agentic workflows locally.

Santhanam said the new Mac Studio M5 Ultra, for example, will provide the “ultimate desktop performance and the ability to run massive AI models” on the device.

Under the hood of the M6 Mac Mini

The new Mac Mini will feature Apple’s M6 chip, which is built using a 2-nanometer process. This will boast a 12-core GPU, two more than the M5, along with a neural accelerator in each core.

Using this design, Apple said this will deliver a near 30% increase in GPU performance for AI compared to the M5, and more than eight-times compared to the M1.

“The super cores blaze through single-threaded workloads, the performance cores use less power and join the super cores to run demanding multithreaded workloads, and the efficiency cores handle everyday background tasks — all with industry-leading performance per watt,” Apple said in a blog post.

Elsewhere, the Mac Mini M6 will support up to 32GB of unified memory, along with 170GB/s of unified memory bandwidth. The firm said this marks a 10% increase over the M5 and will help users run LLMs on the device more efficiently.

The M5 Pro Mac Mini

Alongside the new M6 range, Apple also unveiled the Mac Mini equipped with an M5 Pro chip. This will boast a CPU ranging up to 18-cores alongside a 20-core GPU, and is catered specifically for complex tasks such as 3D design, VFX, and game development.

The device also supports up to 64GB of unified memory, alongside 307GB/s of unified memory bandwidth. Once again, Apple touted the device's capabilities in terms of local AI orchestration.

"With industry-leading performance per watt, Mac mini with M5 Pro is quiet, efficient, and ideal as an always-on desktop for AI agents or creative workflows," the company said in a blog post.

What to expect with the M5 Ultra Mac Studio

According to Apple, the new Mac Studio featuring the M5 Ultra chip is the “ultimate powerhouse for pro workflows” - again placing a strong emphasis on local AI capabilities.

The M5 boasts a next-gen GPU with up to 80 cores, with each also boasting a neural accelerator to provide up to 4.5-times the GPU performance compared to the M3 Ultra chip, and over six-times that of the M1 Ultra.

“The GPU includes Apple’s latest shader core with second-generation Dynamic Caching, as well as hardware-accelerated mesh shading and third-generation ray tracing, delivering up to 40 percent faster graphics performance than M3 Ultra,” the company said in a blog post.

On the memory front, the new device also boasts significant improvements, featuring up to 512GB of unified memory and 1.2TB/s of unified memory bandwidth.

Apple said this will enable users to “store huge datasets entirely in local memory” and run LLMs ranging in the hundreds of billions of parameters locally.

What to expect with pricing

Customers looking to get their hands on the Mac Mini with M6 or M5 Pro chips can pre-order today (25 August), although the devices will set you back.

In terms of pricing, the Mac Mini with M6 will start at $899 in the US, although users in education can get it for a reduced price of $799. The Mac Mini with M5 Pro, meanwhile, starts at $1,699 and $1,599 for education.

Elsewhere, the Mac Studio with M5 Ultra starts at a whopping $5,499 and $5,099 for education.

Designed for devs

Notably, Apple’s various developer frameworks and tools, such as Core AI, Core ML, and Xcode, will “tap directly” into both chips.

This offers marked advantages in terms of local AI orchestration, the company said.

“With these frameworks and new chips, developers can run and fine-tune large AI models locally on their Mac,” Apple said in a blog post.

“Apple’s developer tools and frameworks automatically optimize performance across the CPU, GPU, and Neural Engine, and give developers the ability to use Apple Foundation Models, App Intents to tap into Apple Intelligence features, or their own proprietary AI models to build and run powerful AI workloads entirely on device.”

Another pint-sized powerhouse

Apple’s pint-sized powerhouse, the Mac Mini, is back with an M6 chip – and it comes at a time when mini PCs are having their moment.

Smaller players from China such as Geekom and Beelink are finding more and more markets with their innovative and powerful desktop boxes, but they will struggle to match the pulling power of Apple.

And they won’t be anywhere near as adorable as the Mac Mini.

FOLLOW US ON SOCIAL MEDIA

British tech workers are using AI for 35 tasks per day across their teams, new research suggests. While that works out to just a handful of prompts for most employees per day, it's starting to build up to serious electricity usage.

That's according to Uswitch Business Energy, which found the average workplace team generates 24 prompts per day, including writing emails, automating admin tasks and analysing data.

But that's across the whole team, not per worker. Instead, Uswitch combined its data with YouGov research for some back-of-the-envelope extrapolations.

Survey data shows 32% of British workers use AI at work, and with an average team size of 12, and an estimated average AI prompt generation of two per day, that adds up to an estimated 22 million AI prompts every working day.

That sounds like a lot, but there's just shy of 26 million full-time workers in the UK, and not all of them are sitting at computers.

IT and telecoms workers account the most prompts, at 35 per team, followed by finance at 29, Education at 22, and healthcare at 21, the report found.

Workers aged 25-34 had an average of 30 AI-assisted tasks on their teams each day, versus 27 for younger workers and just 18 for those aged over 55.

Are energy concerns warranted?

Uswitch's energy division warned that the shift to AI is eating up a fair amount of electricity.

The switching company points to Google's own estimate that a Gemini text prompt uses 0.24 watt-hours (Wh) of energy, meaning British employees are using 5.28 MWh of electricity each and every work day, or about 1.37 gigawatt-hours (GWh) each year.

That works out to enough electricity to power 508 British homes annually. Of course, that energy use is over in a Google data center, rather than at point of use.

"AI has rapidly evolved from an emerging technology into an everyday business tool," said Ben Gallizzi, energy expert at Uswitch. "As more organizations adopt AI to support tasks such as content creation, customer service and administration, the cumulative energy demand from these tools will continue to grow."

Developing and training a model is an energy hungry process, with estimates that training current cutting edge models uses between 20-25MW of power over three months, or between 43 GWh and 54 GWh in total.

However, research by MIT Technology Review suggests that those massive figures make up a small amount of the total energy use of AI.

As much as 90% of its footprint lies in inference, with small text based queries, image generation, and the rest adding up when used by so many people.

"Although each prompt uses a relatively small amount of energy, the scale of adoption means businesses are collectively generating billions of AI interactions every week," said Gallizzi, although the billions would be globally, not just in the UK.

Indeed, OpenAI said last year that it was seeing 2.5 billion queries on ChatGPT each day.

Gallizzi added: "As organizations continue to embrace AI, understanding the impact this could have on future electricity demand will become increasingly important."

Time to track

Uswitch suggested businesses need to be aware of these figures, just as they track energy use from more mundane aspects of their offices. As noted, this usage will be off-site in a data center run by AI companies, so it's more about considering overall usage and impact than direct costs.

"Businesses already carefully monitor energy use from equipment, heating and lighting," Gallizzi said. "As AI becomes more deeply embedded in workplace processes, organizations may also need to consider how digital tools contribute to their overall energy footprint."

Wanton AI use could be reined in — not by electricity footprint concerns, but by rising prices and shifting billing models. That's led to companies pushing to end so-called "tokenmaxxxing" and encouraging more sensible use of AI credits.

FOLLOW US ON SOCIAL MEDIA

Enterprises are falling into a trap of using AI when it’s not necessarily needed, and it’s costing them dearly, according to Roger Lee, AVP of solutions consulting at Appian.

Speaking to ITPro, Lee said the hype surrounding AI - particularly agents - is pushing some IT leaders to apply the technology in an ‘AI for the sake of AI’ type approach.

“I think a lot of organizations and a lot of senior executives and staff within organizations have been told to ‘do AI’,” he said.

“So rather than just pushing AI for the sake of AI, look at what the decisions are we’re trying to improve, what data we are using, and where does that human oversight sit.”

Lee’s comments come after Accenture told staff to stop using AI for unnecessary tasks amid surging costs. As adoption of the technology accelerates, he told ITPro that enterprises need to have a serious conversation on how they integrate AI tools.

Agents vs business rules

Pursuing the ‘AI for the sake of AI’ approach means that many businesses aren’t just incurring huge costs, but adding needless complexity to processes that were already streamlined prior to the generative AI race.

“The key thing is really trying to identify the right place for AI versus just trying to use AI to solve everything,” he said.

“In many cases, if it’s straightforward automation, if there’s a business rule where you’ve got a definitive yes or no answer, then sometimes business rules and workflows that result in a definitive answer are the right way to go.”

Admittedly, Lee said there are instances where agents are necessary, especially with regard to research or complex problem solving tasks. Acknowledging that this is a case of different strokes is crucial, however.

“I take an example from insurance with an automation claim, where you wanted to do some research and do some work, then an AI agent might be the right solution for that,” he said.

“But it's really understanding the right place to use a role versus an AI agent, and the impact that that can give.”

This is particularly relevant in regulated industries, Lee said, where margins of error can have huge ramifications.

Agents are highly effective in terms of dynamic reasoning, he noted, but in cases where businesses need a “really definitive zero error response” then IT leaders need to know where and when to automate.

“Regulated industries is another area where humans in the loop are really important,” he said.

“It's what makes processes accountable. So to ensure that not everything is automated, where you have things that are exceptions or outliers, that you absolutely have humans in the loop to respond to those.”

Shoehorning AI into processes

Enterprises have found themselves shoehorning AI into areas and processes that aren’t applicable, research from Gartner shows. Moreover, they’re expecting near-immediate returns with the technology.

Ironically, a key factor behind this lies in the desire to deliver returns on investment with AI. Yet by doing so they’re less likely to unlock any benefits.

Speaking at the time, Melanie Freeze, director of research at Gartner, said that many AI initiatives are “overly ambitious or poorly scoped” and don’t align with “real operational needs”.

Ultimately, Lee told ITPro that as with any adoption process, AI should still be viewed as a “business pull rather than technology push” approach.

“I think that the whole value that you can get from AI within an organization is all around what the business needs, rather than what the technology can push to the organisation,” he said.

“First and foremost, what problems are we trying to solve, rather than where can we use AI, and that changes the perspective on what we're trying to do,” Lee added.

FOLLOW US ON SOCIAL MEDIA

The government has outlined proposals for new powers under the Cyber Security and Resilience Bill aimed at boosting cyber defences for the UK's essential services.

With hostile cyber attacks and sabotage on essential services like energy, water, transport, and healthcare on the rise globally, vendors and companies with connections to hostile states pose a 'growing and serious' threat to the UK’s security.

The idea is for the government to be able to step in when essential service providers are looking to buy equipment or technology from suppliers that could pose a critical national security risk.

"These new powers mean we can act before a threat materializes, not just after the damage is done. By working together with industry, we're putting national security at the heart of how essential services choose their suppliers," said cyber security minister Liz Lloyd.

"This is about staying ahead of a growing threat, and giving the public confidence that the everyday services we depend on like water and energy supplies, our transport network and hospitals, are protected."

New cyber safe procurement guidance would be made available for all essential service providers to help them secure their supply chains. Providers will be able to refer themselves for a risk assessment if they're unsure about a vendor – and be given advice on how to mitigate any risks.

Meanwhile, the package would grant the government new legal powers to issue binding directions to essential service providers, including requiring extra security measures, a phased withdrawal, or in the most serious cases, an outright ban on acquiring a vendor or supplier's products or services.

Growing state-backed threats

The proposals come amidst growing concerns over the impact of state-sponsored cybersecurity threats.

Earlier this year, the National Cyber Security Centre (NCSC) warned that the UK is experiencing increased threats from Iranian-backed cyber groups, with a heightened risk of indirect threats for organizations with a presence, or supply chains, in the Middle East.

Government figures indicate that an outage caused by a cyber attack on London and Southeast England's electricity networks could potentially cost the economy up to £442 billion over the following five years.

Just this week, the vulnerability of key UK infrastructure was brought into focus by the news that Iranian-linked hackers were been able to knock a small power plant offline for four days.

The attack, which targeted internet-exposed programmable logic controllers (PLCs), is believed to be the first of its kind in the UK.

The amendments to the bill were laid before Parliament on Monday, ahead of Lords Committee stage scrutiny, which is set to begin in September.

FOLLOW US ON SOCIAL MEDIA

US lawmakers are pushing back against the Trump administration's decision to reduce headcount at the Cybersecurity and Infrastructure Security Agency (CISA).

Five Democrats have written to the Government Accountability Office (GAO), calling on it to look into the effects of the cuts on the agency's ability to protect critical infrastructure and respond to evolving cyber threats.

The group has requested an investigation into how CISA’s workforce has changed over the past five years, including its size, composition, geographic distribution and number of contractors.

They also want the GAO to identify which programs have been affected by the cuts, what the effects have been, what data CISA collects and analyzes for workforce planning and whether it uses that data to align resources with agency priorities.

"At precisely the moment when our adversaries are accelerating attacks against critical infrastructure, the Department of Homeland Security's Cybersecurity and Infrastructure Security Agency (CISA), the nation's lead civilian cyber defense agency, has lost nearly one-third of its workforce, raising serious concerns about the agency's ability to fulfil its mission," they wrote.

"Little is known about the impact of these workforce reductions on CISA’s programs and services or what processes and plans the agency has in place to ensure the agency is hiring the right people to address lost skill sets and meet mission demands."

CISA cuts run deep

The latest changes were announced in June as part of the 2027 budget proposal. CISA is set for a cut in funding of around $700 million - with the White House claiming that it's been functioning “as a hub in the Censorship Industrial Complex, conspiring against the First Amendment rights”.

The cuts would refocus CISA on its core mission – Federal network defense and coordinating with critical infrastructure partners – while eliminating waste by consolidating redundant security advisors and programs. Around 900 jobs are expected to go.

CISA lost nearly 1,000 employees, or about a third of its workforce, in the first half of 2025 - although the agency has since made moves to try and boost staffing again.

It's also seen cutbacks to its work on election security. And all this has come as its work has expanded more than ever, thanks to an accelerating threat environment.

Gene Moody, field CTO at Action1, said the cuts show a lack of awareness over the critical work conducted by the agency.

"In many respects, vulnerability analysis is further behind the curve than it has ever been. And it is being attacked by its own governing bodies out of ignorance and misunderstanding. This is an existential threat to cybersecurity," Moody commented.

"When analysts and programs are overloaded, the ecosystem gets noisier: important vulnerabilities compete with an enormous amount of lower-value information, prioritization becomes harder, and defenders have less confidence about what actually deserves immediate attention."

FOLLOW US ON SOCIAL MEDIA

Semiconductor revenue is set to almost double this year as the AI sector continues to reshape the market – and drive up prices for in-demand components.

New figures from Gartner project global semiconductor revenue growth to top 92% this year, reaching $1.6 trillion over last year's $809 trillion. In 2027, that’s expected to climb further to $1.9 trillion, the company said.

"The semiconductor industry is entering a fundamentally new phase of growth," said Ben Lee, Director Analyst at Gartner. "The pace of industry expansion is accelerating dramatically, driven by sustained investment in AI infrastructure and a stronger-than-anticipated memory pricing cycle."

That's good news for those manufacturing semiconductors, but the rise of AI – and the surge in demand that it's sparked for components – has led to price rises across a range of industries.

In the cloud market, a host of providers have issued price hike warnings. Meanwhile, consumer electronics are being hit hard by rising hardware costs. Apple recently warned that prices will rise this year, for example, while PC shipments have hit record lows.

Memory and beyond

Semiconductor gains are largely driven by memory revenue, with prices for the key component skyrocketing because of AI demand. Some predictions suggest that AI will consume a fifth of all DRAM capacity in 2026.

That shortage is sending prices skyrocketing and already impacting businesses' plans to upgrade tech.

This trend is expected to continue, according to Gartner. The firm predicted DRAM revenue would increase 246.6% this year, with NAND flash revenue up by 371.9% – and that’s despite additional capacity being made available.

Memory now makes up 54% of total semiconductor revenue, a huge leap from 27% last year. Revenue for memory will top $837bn this year and $1trn next year, Gartner predicts.

"AI infrastructure has fundamentally changed the dynamics of the memory market,” said Shrish Pant, Director Analyst at Gartner.

“While pricing expansion is accelerating growth in 2026, continued AI infrastructure deployments, higher memory content per AI server and sustained demand for high-bandwidth memory (HBM) will support memory revenue growth through 2027 and beyond."

While semiconductor revenue growth is dominated by memory price rises and demand, Gartner noted that the AI roll-out is benefiting all aspects of the industry, as AI data centers need CPUs, networking, power management, and more.

"AI is expanding the semiconductor opportunity across the entire infrastructure stack rather than benefiting a single device category," said Lee.

Gartner said semiconductor revenue growth excluding memory would grow from $589 billion in 2025 to $718 billion in 2026, an increase of 21.9% – and reach $864 billion in 2027.

AI data centers

Notably, the mass buildout of data centers is changing how semiconductors are used and where they end up, according to Gartner.

"As AI infrastructure scales, it is not only reshaping investment across the semiconductor ecosystem but also redefining the industry’s application mix," said Lee, saying Gartner sees a "a structural shift in where semiconductor value is created and how demand is evolving across the industry.”

Gartner predicted that the AI data center ecosystem would make up over a third (36.5%) of semiconductor revenue this year, climbing to more than half (53%) by 2030.

FOLLOW US ON SOCIAL MEDIA

Freeths has announced the appointment of John Jones as its new chief technology officer (CTO), in a move the law firm said will strengthen its leadership bench and growth ambitions.

Jones will work across Freeths’ twelve offices located across the UK and will oversee the shaping and delivery of its technology strategy.

His remit will include ensuring the business has secure, resilient, and future-ready technology platforms that support employees, improve the client experience, and enable operational efficiency.

In an announcement, Freeths national managing partner Karl Jansen said the appointment comes at an “important point” in the company’s ongoing growth and development.

“Technology is central to how we support our people, serve our clients and operate as a modern law firm,” he commented.

“John’s depth of experience, combined with his pragmatic and inclusive leadership style, will be instrumental as we continue to build scalable, resilient and innovative technology foundations across the firm.”

Headquartered in Nottingham, Freeths is a UK commercial law firm serving private and public sector clients, including organizations such as Centrica, Aldi, Mercedes-Benz UK, Tarmac, Experian, and Harworth.

Jones joins the business with more than 30 years’ experience across technology and professional services. Most recently, he served as CTO at Forvis Mazars and previously held senior tech roles at BDO.

He began his career as a software developer working on business-critical applications before moving into senior technology leadership positions within the professional services sector.

Jones’ experience includes working with enterprise platforms such as Microsoft, Workday, and legal-specific technologies, while his expertise spans data, digital skills, large-scale transformation programs, as well as the safe and effective use of emerging technologies such as AI.

Commenting on his new role, Jones said he was impressed by Freeths’ existing technology platforms, as well as the clarity of its plans for the future.

“Professional services, and the legal sector in particular, is going through significant change, and technology has a vital role to play in enabling better outcomes for clients and colleagues,” he explained.

“I’m excited to be joining Freeths and to be working closely with teams across the business to support the next stage of the firm’s journey.”

FOLLOW US ON SOCIAL MEDIA

While tokenmaxxing has taken the tech industry by storm in 2026, it’s time enterprises start ‘storagemaxxing’ to avoid being hit with huge infrastructure costs.

That’s according to David Boland, VP of cloud strategy at Wasabi. Speaking to ITPro, Boland believes that a confluence of industry trends means IT leaders need to sharpen their focus on storage optimization.

Hardware supply chain issues and rising prices are an area of acute concern for IT leaders in 2026 and are forcing some to alter strategies to compensate for growing costs, he said.

“We all know that the supply chain issue for storage is big news,” Boland told ITPro. “It’s really hard for organizations, especially if they don’t have deep pockets or are large, well-known organizations; it’s always difficult for them to get the hardware that they need.”

The result, according to Boland, is that they “need to look elsewhere” for alternatives or maximize the use of their existing infrastructure.

With AI in the equation, this adds a new layer of complexity to storage optimization, Boland noted. Simply put, there are a “ton of different storage options out there” yet each AI use case has different requirements from a storage perspective.

“Whether it’s inexpensive, long-term storage for compliance and regulatory reasons, or in the training process or the preparation process, [such as] storing checkpoints of training,” he explains. “Each one of these, let’s call them phases, in the AI lifecycle have a different storage need.”

Given the varied storage requirements when it comes to AI data sets, Boland said that some enterprises find their existing infrastructure isn’t compatible.

“Some high performance, expensive storage is not a great fit for your long term archiving, so therefore knowing where to put your data at which phase of that AI pipeline helps you maximize your investment or maximize the amount of money you spend.”

Storagemaxxing explained

Storagemaxxing is essentially a term designed to “give some new life” to traditional storage optimization processes, Boland said, and one that is inspired by various social media ‘maxxing’ trends.

We’ve all seen them online: ‘looksmaxxing’, ‘tokenmaxxing’, even ‘hobbitmaxxing’.

With storagemaxxing, this is primarily about streamlining storage to accommodate for various stages of the AI lifecycle, be that from data collection to training, deployment, and retention.

The aim here is to ensure that data stays “AI-ready, secure, cost-predictable, and free to move between cloud and compute providers as workloads change”, Boland noted in a recent blog post.

Notably, this has a critical impact on both AI and cost efficiency. AI tools need swift access to the data that feeds them. Ensuring seamless access by optimizing storage also reduces the strain placed on infrastructure.

Costly changes

Being “free to move” data between cloud and compute providers is an area Boland said many enterprises have been caught off guard with in recent months.

Many enterprises didn’t anticipate certain costs associated with storage, he said – be that storing data on the wrong platform or in the wrong cloud tier.

This has a direct impact on costs, he said. Even simply accessing that data or moving it can incur further charges due to egress fees. Indeed, Wasabi’s 2026 Cloud Storage Index survey found 48% of storage costs now go toward fees rather than actual capacity.

As ITPro reported at the time, this is severely impacting IT budgets and forcing a rethink of storage strategies.

“Being locked into a storage solution because you can't afford to move that data out, either through egress charges or retrieval fees, is an issue,” he said.

“I see that becoming more of an issue in the AI space. What's nice about hyperscalers is they have a broad range of services that go from GPU to CPU to high-performance storage to low-cost object storage,” Boland added.

“If there comes a point when a customer wants to move that data to another provider for either a better GPU cost or for availability of services, then they get hit with fees that they don't anticipate.”

FOLLOW US ON SOCIAL MEDIA

Tens of thousands of previously-exposed AWS access keys are still active and valid, researchers have warned, with hundreds holding full admin rights.

Truffle Security said its scanners verified 64,024 unique AWS key pairs across 431,875 public findings that surfaced publicly between August 2022 and August this year.

This includes git history, Hugging Face datasets, Docker images, package registries, and CI logs, the company said.

It then re-verified the 10,616 leaked keys with complete credentials and found that 88% still authenticate, including 526 root keys and 242 identity and access management (IAM) users holding AdministratorAccess - which have full admin rights.

"The largest single source is Hugging Face. AWS credentials are the second most common secret type we verify there, with 8,482 unique live keys across 3,394 public datasets," said the firm.

"Hugging Face keys also skew privileged: 17.9% are root, the highest share of any source we track. Most of those datasets are snapshots of public code repackaged for training."

Of the live keys with creation dates, the median age was 1,831 days. Half are over five years old. The oldest was 17.4 years, nearly as old as IAM itself.

"Only 25 keys (0.9%) were created in the last 30 days. Almost none of this population leaked recently. The count has been building for years," Truffle Security said.

"Rotation is the rarer event. Of the keys where we could enumerate the user's access keys, only 13.7% (398 of 2,903) have any newer key alongside the leaked one. The other 86% were never rotated, superseded, or cleaned up."

Of the 7,590 active IAM users, 929 carry AWS's own AWSCompromisedKeyQuarantine policy, and were detected by AWS as exposed and restricted. Of these, 112 carry the original version, which AWS stopped applying in 2023.

Researchers noted that they were flagged at least three years ago and their owners notified by AWS. However, no action was apparently taken and the keys still authenticate.

Tighter access controls

Truffle Security said organizations delete root access keys. One-in-six leaked keys is root, for example, and as such there is no longer any legitimate reason for a root access key to exist in 2026.

They should also sort their IAM keys by age and set a budget alarm, even a small one, to catch cryptomining early - 90.5% of leaked-key accounts have none.

Elsewhere, researchers said security teams should assume that committed means leaked. Nearly half (43%) of keys were sighted more than once across repos, datasets, and images. Deleting the file does not help once it is in a training corpus.

"Any time AWS is aware of exposed keys, we notify the affected customers. We also thoroughly investigate all reports of exposed keys and quickly take any necessary actions, such as applying quarantine policies to minimize risks for customers without disrupting their IT environment," an AWS spokesperson told ITPro.

"As always, customers can contact AWS Support with any questions or concerns about the security of their account."

FOLLOW US ON SOCIAL MEDIA

A small UK power plant was reportedly shut down for four days last month following a cyber attack attributed to Iranian hackers.

The incident appears to have affected a small-scale gas generator, rather than a large power station, and did not affect the wider grid.

According to the Daily Telegraph, the government has contacted the chief executives of power companies following the incident and has written to businesses with advice on the steps they should take to protect themselves.

The attack has been attributed to hackers affiliated to the Iranian government, and is believed to be the first of its kind in the UK.

Graeme Stewart, head of public sector at Check Point, said the incident highlights the potential risks posed by threats to critical infrastructure.

“This marks a grave escalation in the Iran conflict because a hostile state-linked cyber threat has reportedly reached into UK energy infrastructure and caused a physical shutdown lasting four days,” he said.

“That should concern every organization responsible for keeping this country running. For most Brits, the Iran conflict is happening thousands of miles away and cyber warfare probably still conjures up images of stolen passwords, leaked data and companies being held to ransom,” Stewart added.

“The prospect of a hostile state being able to reach into the infrastructure beneath our everyday lives changes that dramatically."

Rising state-sponsored threats

Notably, the incident coincided with a series of attacks on US water supply infrastructure in recent weeks. More than a dozen states were affected in the campaign, which is believed to be linked to Iranian-backed threat groups.

The attacks targeted internet-exposed programmable logic controllers (PLCs), with the attackers remotely changing IP addresses and turning on and setting passwords.

Several water firms were left unable to view connected equipment, and in some cases it was shut down.

The FBI has recommended disconnecting PLCs from the public-facing internet, strictly controlling access to them, tightening up passwords, and reviewing project files running on PLCs for unauthorized changes.

UK infrastructure in the crosshairs

According to the National Cyber Security Centre (NCSC), the UK is experiencing increased threats from Iranian-backed cyber groups.

Earlier this year, the agency issued a warning over heightened risk of indirect threats for organizations with a presence, or supply chains, in the Middle East.

Stewart suggested that the impact of an attack on larger power supply operators or other critical infrastructure firms could be disastrous.

"We have to ask what happens if the next target is bigger, more critical or more deeply connected to the services millions of people rely on. Britain’s critical national infrastructure underpins almost every part of modern life, including electricity, water, transport and communications, and those systems are increasingly digital, interconnected and dependent on one another," he said.

"The question now has to be whether Britain is genuinely ready if something more serious follows.”

A spokesperson for the UK government told ITPro that it's working on an Energy Resilience Strategy. Set for publication later this year, this will outline plans for ensuring the energy system remains stable and secure.

"The UK has a highly resilience energy system. We work closely with the energy sector to protect infrastructure and ensure the highest standards of security," the spokesperson added.

FOLLOW US ON SOCIAL MEDIA

OpenAI CEO Sam Altman believes AI could spark a new wave of small businesses, but the tech industry has a lot of work to do convincing entrepreneurs how the technology can help.

Speaking on the David Senra podcast, Altman suggested AI could help break down traditional barriers for entrepreneurs, particularly in terms of financing and technical capabilities.

“Even if most people don’t want to start really big companies, a lot of people want to start smaller companies, and that has been hard,” he said.

“That has been something that has required a fair amount of privilege, and luck, and resources to be able to do. We are about to see the greatest boom in people starting smaller businesses that we have ever seen,” Altman added. “I think AI is empowering that.”

Altman’s optimistic outlook on this front is reflected in recent research conducted by OpenAI. As ITPro reported in March, a survey by the firm highlighted a strong appetite for AI tools among small businesses across the UK.

Eight-in-ten small businesses across the country now use the technology in daily operations, OpenAI found, although the study warned that uptake varies wildly on a region-by-region basis.

Notably, Altman told Senra that OpenAI and the broader tech industry hasn’t quite emphasized the potential benefits of AI to entrepreneurs and small businesses in recent years.

The focus has largely been on larger organizations and how the technology is impacting operations in these domains.

“For some reason the field, including us, has not talked about that enough, even though we see all the signs of it,” he said. “We have not built enough products to accelerate that, but I think we’re going to see a lot more of that."

Some SMBs are pulling ahead

Altman’s comments come in the wake of recent research from Dell Technologies which found small and medium-sized businesses (SMBs) are increasingly optimistic about the impact of AI.

The study found two-thirds (66%) of UK small businesses view AI as a “route to growth” while more than half (56%) believe the technology will give them a competitive advantage.

Brian Horsburgh, UK small business country manager at Dell Technologies, told ITPro that small businesses have a huge advantage over larger companies in terms of adoption.

Smaller firms are far more agile and typically more flexible in how they integrate the technology in daily operations. This, the study found, is highlighted in the rate of successful adoption projects.

Indeed, larger companies tend to cite a broader range of barriers when it comes to adoption, including lengthy approval processes, fear of failure, or resistance to change across the workforce.

FOLLOW US ON SOCIAL MEDIA

AI represents a prime opportunity for security teams to keep pace with accelerating development timelines, according to GitLab CISO Chaim Mazal. But this will require tweaks to traditional processes.

The use of AI in software development is fast approaching near-universal levels, with figures from Stack Overflow’s 2025 Developer Survey showing 84% of developers have adopted the technology.

While AI is helping speed up code production, streamline development lifecycles, and drive productivity, Mazal told ITPro it’s also creating new security risks.

“Businesses have decided that in order to be successful in this next wave of the journey, AI has to be leveraged and teams have to go full bore,” he said.

“I think in tandem and parallel, security teams have been working on figuring out how they can build appropriate solutions to be able to enable their organizations to do that.”

Mazal noted that enterprises are still working around the “same practices and principles that we’ve historically followed” - robust testing, reviews, and security checks.

But with the advent of agentic AI and the pace of change, there’s a risk that blind spots could emerge somewhere in the process.

Indeed, it’s an issue that GitLab recently highlighted in a study. Analysis from the company in June warned that eight-in-ten organizations are adopting AI tools faster than they can develop policies to govern them.

The influx of AI-generated code is also proving troublesome for enterprises, with a study from Tricentis showing that roughly 60% of organizations have shipped untested code.

The result here is that security teams now face a confluence of challenges. Potentially dangerous AI-generated code and increasingly fast development timelines are placing huge pressure on teams to ensure products ship in both a timely and secure manner.

An engineering-first approach

Keeping pace with AI-powered development will mean security teams essentially have to start operating more like their counterparts in development in an ‘engineering-first’ approach.

This is an operating model that takes core engineering principles such as automated testing and CI/CD pipeline techniques and embeds them within security processes. Simply put, rather than acting like manual auditors and policy enforcers, they play a more involved role in development processes.

Moreover, they build parallel products designed to support development from the get-go. Fundamentally, this is about being a “first-level contributor” in software development, Mazal said.

“As our engineering teams move fast, having the security team have the ability to contribute code, make iterative adjustments, and be part and parcel with the development process is key to our success,” he told ITPro.

“Being able to move fast, iterate fast, and not just be consultative in nature, and being able to really roll up our sleeves and work alongside the rest of the teams to build that secure path and the secure guardrails is something I think most modern security teams don’t have the luxury of foregoing.”

When it comes to AI, what’s good for the goose is good for the gander. Security teams should view the technology as a means to keep up with development processes – as with developers, teams need to automate where they can.

Humans in the loop

Ensuring humans remain in the loop throughout the development and testing process will still be vital moving forward, according to Mazal. The exact scope of human involvement is still up for debate, however, and the long-term goal is to automate as much as possible in the safest way

“I think most organizations who make commercial-level software want to get to a place where there are only humans in the loop in places that it’s absolutely necessary, based on the risk to the organization,” he said.

“They’d like to be able to have the overwhelming majority of their code auto-generated and auto-reviewed.”

Mazal said this has the potential to pose new challenges for developers, testers, and security teams alike. Fully-automated pipelines are still evolving, and as ITPro recently reported, some enterprises are exploring the concept of ‘dark testing factories’ to ramp up automation with little human input.

“Collectively we’re still trying to figure out as an industry what those best practices are, but we’re definitely making very big leaps and bounds in very short periods of time.”

GitLab itself is ramping up the use of agents in a software factory approach, Mazal noted, focusing intently on a security-conscious strategy. Agents need strict guardrails, and enterprises need to implement certain restrictions from the get-go.

“How we’re thinking about this at GitLab from a software factory perspective is that you have the ability from the very initial inception of design to bake in those security guardrails and requirements,” he explained.

“You can go ahead and create and craft them and define them based on the data sensitivity and requirements that you have, and you can uniformly implement governance controls across the board.”

These processes will naturally evolve as requirements change, Mazal said, and teams will “continuously make adjustments along the way”.

FOLLOW US ON SOCIAL MEDIA

AMD has boosted rack efficiency four-fold since 2024, and the company claims it’s on track to upping it by 20 times by the end of the decade.

The chipmaker revealed it recently exceeded its 2026 target of a three-fold improvement in rack-scale energy efficiency for AI training and inference.

Two 2030 AMD racks are now expected to deliver the same compute as roughly 570 racks did in 2024, enabling a 20x reduction in use-phase electricity and a 28x reduction in carbon intensity.

Alternatively, efficiency gains are expected to enable 20-times more compute, measured in floating point operations per second (FLOPs) per watt, using the same amount of energy.

AMD said energy improved energy efficiency delivers a range of environmental benefits, but also in helping to improve AI performance without increasing power, improving total cost of ownership, and helping customers scale faster.

“The next wave of AI efficiency will depend on tighter co-optimization across compute silicon, memory, interconnects, software and rack-scale system design," said Sam Naffziger, senior vice president and corporate fellow at AMD.

"Our estimated 4x improvement through 2026 reflects the strength of our approach and the progress AMD is making across the full system, putting us ahead of our projected pace toward the 2030 goal.”

The improvements in performance come down to compute capability, memory bandwidth, and interconnect bandwidth, according to AMD.

Advanced process technology and architectural improvements help increase floating-point compute performance per watt, advanced architectures and memory integration improve bandwidth, and high-speed interconnects and tighter integration increase the network bandwidths.

High-bandwidth memory, larger caches, and tighter integration between memory and compute can reduce unnecessary data movement, which uses energy. Meanwhile, high-speed scale-up interconnects help larger systems reduce bottlenecks and share data more efficiently.

AMD’s ROCm software driving gains

Elsewhere, AMD claimed its ROCm software stack and open standards help developers and organizations deploy AI workloads more efficiently on AMD platforms, with open ecosystems also supporting broader optimization across AI and high-performance computing (HPC) deployments.

"The projected rack-scale efficiency gains AMD has set forth can enable AI and HPC customers to achieve data center-level gains in power and cooling infrastructure, which can be limiting factors to increasing compute performance," said the firm.

"Better energy efficiency can help reduce operational electricity use, improve total cost of ownership, and support more sustainable AI and HPC growth."

AMD has also published its 2025-26 Corporate Responsibility Report, detailing its progress across product efficiency, operations, supply chain responsibility, and community impact.

It reduced operational greenhouse gas emissions by 30% from 2020 to 2025, sourced 58% of global AMD electricity from renewable sources and audited 99% of its manufacturing supplier factories for responsible business practices.

"We aim to cut operational greenhouse gas (GHG) emissions in half from 2020 to 2030 as we expand our use of renewable energy. In 2025, 100% of AMD manufacturing suppliers maintained public GHG reduction targets and 82% sourced renewable energy, said AMD chair and chief executive officer Lisa Su.

"The AI era is still in its early stages. Working closely with industry, government and academia, we will keep building the high-performance, energy-efficient compute foundation it requires."

FOLLOW US ON SOCIAL MEDIA

AI cloud firm Nebius is leasing high-density capacity at Vantage Data Centers' Newport campus, marking the first commercial capacity commitment in the South Wales AI Growth Zone.

Under the agreement, Nebius will lease Nvidia-powered capacity at Nebius's CWL1 facility to support AI training, inference, agentic AI and enterprise AI workloads, and serving enterprises, researchers, start-ups and public sector organizations.

“This is an important milestone for South Wales and for the UK’s AI infrastructure ambitions,” said David Howson, president, EMEA, Vantage Data Centers.

“Nebius is scaling AI cloud infrastructure in the UK at a time when demand for domestic compute capacity continues to accelerate. Vantage’s Newport campus provides the scale, connectivity and operational excellence needed to support that growth, and we’re proud to help establish South Wales as a leading destination for AI investment.”

One of Europe’s largest data center campuses, Nebius's CWL1 site opened in 2010. It runs on 100% certified renewable energy, with its latest facilities designed to minimize water consumption through a closed-loop cooling system that recirculates water rather than evaporating it.

The facility also uses air-cooled chillers and free cooling when ambient conditions permit.

The increased capacity forms part of Nebius’ broader UK expansion and commitment to scale domestic AI compute capacity.

In June, the company announced around £1.7 billion of committed capacity build-out across four UK sites, including the delivery of more than 1GW of AI-ready capacity across Newport, Bridgend, and Bro Tathan.

“The UK is one of the places where AI is being built, deployed and adopted by startups, enterprises and the public sector,” said Gary Tierney, general manager for EMEA at Nebius.

“Vantage’s Newport campus gives us a strong foundation to expand access to NVIDIA-powered AI infrastructure in the UK and support customers building the next generation of AI applications.”

AI Growth Zones take shape

The buildout is based on Nvidia's latest full-stack, end-to-end DSX AI factory reference design, and supports the UK government’s AI Opportunities Action Plan.

South Wales was designated as one of the UK's UK AI Growth Zones in November last year. At the time, the government said it could create more than 5,000 new jobs for local communities.

It's also home to the UK’s largest cluster of semiconductor businesses.

The AI Growth Zone stretches along the M4 corridor from Newport to Bridgend has the potential to harness over 1GW by early 2030s. 

“Today’s announcement marks an important milestone for the South Wales AI Growth Zone and demonstrates Wales’ growing reputation as a strategic location for the infrastructure that will power the AI economy,” said Adam Price, Welsh cabinet minister for enterprise, connectivity and energy.

"Nebius and Vantage’s investment will help unlock new economic opportunities, support highly skilled jobs and reinforce Wales’ position as a leading destination for advanced digital investment."

The move by Nebius comes hot on the heels of AI Growth Zone announcements elsewhere across the UK this week. The Lanarkshire growth zone in Scotland, for example, secured £300 million in investment to ramp up data center development, with Dell Technologies pledging its support for the site.

FOLLOW US ON SOCIAL MEDIA

Visitors to Black Hat/DEF CON earlier this month were targeted after the event by cyber criminals posing as a well-known crypto media executive.

The X account @HartmansDoeke sent a direct message to one Huntress security researcher claiming to be CoinDesk's VP and head of marketing, asking for help with an upcoming conference.

While the researcher cottoned on to the scam immediately, they carried on engaging with the scammer to check out the tactics they were using. This involved a Google Doc featuring a custom Google Apps Script sidebar designed to guide them through the execution of malware.

The document asked the potential victim to enter an 'encryption key' – supplied by the actor in direct messages – which appeared to fail when entered. The sidebar provided two follow-on options: ClickFix-style instructions and a download option, both intended to download and execute malicious code.

Mac users were served an infostealer targeting browser passwords, crypto wallets and even private Notes app data. Meanwhile, Windows users were served a remote access trojan, a fake crypto wallet implant, and a network-intercepting proxy delivered via an installer signed with what appears to be a stolen certificate.

When the researcher didn't fall for the malicious Google Doc, the threat actor followed up the next day with a second malicious document.

This masqueraded as a Dropbox DocSend share and led to a counterfeit DocSend installer that delivered AMOS stealer to macOS users and NetSupport RAT, a Ledger wallet implant, and a traffic-intercepting proxy to those on Windows.

"Taken together, the two lures show how the threat actor used familiar platforms to build credibility and keep the target engaged," Huntress said. "By combining social media DMs with trusted document and file-sharing services, the actor created a legitimate-looking workflow designed to trick targets into running the malware."

Conference attendees urged to remain vigilant

According to Huntress, the researcher was just one of many to be targeted.

"Large industry events like Black Hat and DEF CON create a target-rich environment for bad actors, with attendees exchanging new contacts, documents, invitations, and follow-up plans," Huntress said.

"Attackers are using this activity to make malicious outreach look like just another routine post-conference interaction."

Anybody who's interacted with a lure like this is advised to isolate the system from the network, collect any relevant forensic evidence, and consider reimaging the system.

They should assume that credentials on the system have been compromised and revoke active sessions, reset passwords, and rotate API keys or any other secrets that may reside on the system – and also review any cryptocurrency wallets.

While visitors to a security conference might not seem like the most obvious victims, this wasn't the only attempt to scam this year's DEF CON attendees.

One passenger on a Delta flight out of Las Vegas attempted to jam in-flight Wi-Fi and broadcast a rogue network designed to look like the airline’s service, in an apparent phishing attempt.

FOLLOW US ON SOCIAL MEDIA

Exclaimer has announced the launch of MSP Connect, a new partner program designed to help managed service providers (MSPs) provision, manage, and scale email signature services across customer environments.

Available immediately, the initiative brings consumption-based billing, centralized self-service provisioning, and dedicated channel support together with integrations for ConnectWise, Kaseya, and Microsoft Marketplace.

Exclaimer said the program has been designed to address the operational challenges MSPs face when managing email signatures across Microsoft 365 and Google Workspace environments, where updates can otherwise require manual, user-by-user administration.

With MSP Connect, billing adjusts according to actual customer usage, allowing MSPs’ costs to increase or decrease as customers add or remove users.

MSPs will also have access to a redesigned partner portal that allows them to provision and manage signatures across multiple customer environments from a single location, while native integrations with PSA platforms aim to reduce administrative work.

In an announcement, Exclaimer chief operating officer Jim Turner said MSP Connect was developed in response to feedback from partners around the ease of use and billing model of the program’s predecessor.

"MSPs have been clear about what they need from us: a platform that is easy to provision at scale, a commercial model that flexes with their customers and commercial support that helps them grow," he explained.

“MSP Connect is the reset. It removes unnecessary steps, gives partners more control through self-service, and aligns billing with what their customers use."

Strengthening MSP revenue generation

Exclaimer’s MSP Connect launch marks the next phase of its channel-first growth strategy as the email signature management specialist looks to strengthen its partner network.

Partners already account for 30% of the firm’s annual recurring revenue, while its global channel ecosystem includes around 5,000 partners.

The company said MSP Connect will now enable partners to add email signature management to existing Microsoft 365 and managed services relationships without introducing a complex deployment or support burden.

MSPs can generate license margin, bundle Exclaimer’s technology into wider managed services packages, as well as create additional revenue through services such as signature design, deployment, and ongoing management.

"An MSP is a business within a business. It needs to manage hundreds of customer environments without adding operational drag," commented Louise Taylor, vice president of channel at Exclaimer. "If getting one customer live requires a long sales process or a services project, the economics stop working."

Reducing friction for MSPs

Exclaimer’s new program includes not-for-resale (NFR) licensing, dedicated channel account management, as well as co-brandable partner enablement materials.

The vendor said its platform can be deployed by MSPs in under an hour depending on the customer environment, using directory data to populate approved signature attributes across Microsoft 365 and Google Workspace without requiring any heavy integration project or mandatory professional services.

Additionally, MSPs can build additional services around the platform, including signature template design and template catalogue creation, while their customers’ marketing teams can utilize approved campaign banners, engagement analytics, and rule-based signature content.

FOLLOW US ON SOCIAL MEDIA

Many IT leaders are “avid gamers”, according to new research from Logitech, and this increasingly shapes how they make business and hiring decisions.

While gaming has traditionally been viewed as a hobby, the study noted that there are a range of transferable workplace skills. 98.8% of IT decision makers said that gaming directly affects their job performance and competency, for example.

IT leaders said gaming benefits them in areas such as strategic thinking (65.7%), problem solving (65.5%), and performing under pressure (63.5%), according to Logitech.

"Savvy enterprise tech buyers don't just shop for features – they stress-test systems and demand friction-free workflows,” said Henry Levak, general manager for Logitech’s team workspace solutions division.

“Managing complex IT infrastructure requires quick thinking, constant adaptation, and staying calm under pressure—the same skills gamers use every day.”

Gamers get the jobs

Notably, Logitech found 82% of IT leaders are more likely to consider job candidates who list gaming experience on their resumes.

A key factor here, the study noted, is that it points to skills such as “multiplayer coordination and strategic execution” - both of which are viewed as valuable real-world attributes.

This trend shows no signs of slowing down either, according to Logitech.

Looking ahead, more than 90% of IT leaders said they expect technology leadership to become “increasingly influenced by gaming standards” – particularly with regard to team performance.

"The data validates a shift many tech executives already recognize: modern players are essentially the architects of the future technical labor pool," noted Robin Piispanen, general manager at Logitech Gaming.

"The cognitive flexibility, rapid response times, and analytical capabilities honed in gaming environments are proving to be significant professional assets in the contemporary office."

Transferable skills

Logitech’s study isn’t the first to highlight the benefits of gaming. Previous research suggests that gaming has a direct benefit in terms of building ‘soft skills’ such as communication, teamwork, and time management.

In a study conducted by the University of Surrey, researchers found that online gaming also helped improve job prospects.

“Our study found that online gaming behaviour varied between different job categories, allowing the participants to gain different soft skills,” the study noted.

“The soft skills gained could assist gamers with training that leads to a particular career path.”

Some industries have begun looking to areas such as esports as a means to attract digital-savvy talent, including the UK Armed Forces.

As ITPro previously reported, the Ministry of Defence (MoD) recognized esports as an official sport. Some British Army and Royal Air Force (RAF) sites even have dedicated esports facilities for personnel.

FOLLOW US ON SOCIAL MEDIA

GitHub has revealed the root cause behind a nearly eight-hour long outage on Monday: a misconfigured policy that triggered a cascade of failures.

On Monday, GitHub users began struggling to use the service around 1:28pm UTC, reporting issues with API requests, Actions, Webhooks, Pull Requests, and GitHub Copilot. At the time, GitHub said it was seeing error rates between 20% to 50% for different tasks, with other services also impacted.

"At peak, web/API error rates were approximately 20%, while archive and raw-content downloads reached approximately 50%," GitHub said via an update on its status report.

"SAML/OIDC authentication, SCIM, and Team Sync were also affected, as well as Actions workflows in GHEC with Data Residency that depend on public workflow step definitions hosted on GitHub.com."

Within a few minutes the problem was spotted, but it took a total of seven hours and 47 minutes for the issues to be totally resolved, the company said.

"Most services recovered by 16:36 UTC as our Central US datacenter recovered; Actions was degraded until approximately 18:03 UTC; and Copilot Token Service fully recovered by 21:02."

What happened with the GitHub Outage?

After an investigation, GitHub said the issue was due to “network saturation” of load balancers due to new peak traffic in the Central US hosting region.

"Originally this was caused by an Istio sidecar pod reaching its concurrency limits and failing to auto scale correctly because of a misconfigured policy that watched host service but not sidecar limits,” the company explained in a post-mortem report.

That failure cascaded, the post noted, with four HAProxy nodes exhausting their flow limits, degrading the gateway authentication path — all of which led to further authentication latency and failures, the GitHub said.

"The problem was worsened by optimistic retry logic which overloaded internal load balancers," the post added. "Pausing HAProxy on those nodes simultaneously produced immediate broad recovery."

Slow recovery

GitHub immediately moved some traffic away from the failing Central US data center location to Northern Virginia to give time to debug and resolve the issues. But Northern Virginia was quickly flooded with traffic too because of a retry bug.

That was pinned on "client retry behaviour" – in other words, client software kept trying to make it work over and over again, leading to failed requests and amplified load.

GitHub managed to temporarily block the "retry storm" but still saw a huge increase in Copilot authentication.

"Copilot Token Service traffic increased from a normal 7–9K RPS to 70–100K RPS. Reducing gateway authentication retries and blocking retry-triggering responses stabilized Copilot Token Service and completed recovery," the post noted.

External factors exacerbated the issue, according to GitHub, including a number of “scraping attacks on codeload endpoints”.

What's next?

GitHub said it had solutions lined up to avoid a similar outage. That includes fixing the initial autoscaling policies and auditing Istio request, concurrency ,and scaling limits, in order to avoid the same flaw recurring.

It also plans to review "retry limits and backoff behaviour across gateways and clients," and address the VS Code retry behaviour that led to the massive leap in Copilot token traffic.

More widely, GitHub plans to improve load balancer capacity monitoring and failover safeguards.

FOLLOW US ON SOCIAL MEDIA

Hackers are using AI-generated exploitation scripts disguised as legitimate monitoring tools to hack Siemens programmable logic controllers.

An advisory from the US National Security Agency (NSA), the FBI, and other agencies warns of an active threat against the Siemens controllers, which are widely used in critical infrastructure sectors such as manufacturing, energy, and agriculture.

The threat actors use internet scanning services to find internet-exposed PLCs running outdated software, or that are otherwise poorly protected.

"This is not a theoretical risk — it is an active threat," the Cybersecurity and Infrastructure Security Agency (CISA) warned.

"Depending on the specific circumstances, exploitation of poorly protected PLCs could lead to disruption of critical industrial processes, safety incidents, downtime or equipment damage, compromise of sensitive data, compliance violations, and cascading impacts across interconnected systems."

Using AI to generate exploit scripts is a new tactic, CISA noted. This helps dramatically reduce the time, effort, and technical expertise required to develop working industrial control system exploitation scripts and malicious tools.

The tactic also allows hackers to exploit additional attack vectors at speed and adapt to defensive measures by collecting public information about vulnerabilities and weaknesses, finding exposed and exploitable PLCs and using AI-generated scripts to act on that information.

"If PLCs are exposed to the internet, they are at high risk for exploitation," CISA said.

Siemens S7 Series users urged to act now

Security agencies have urged all owners and operators of operational technology (OT) systems using Siemens S7 Series and other PLC devices to make sure their systems are properly protected.

This includes implementing applicable security patches and updates, ensuring isolation from the internet wherever possible, and having strong access controls and security tooling to monitor environments for malicious activity.

Andrew Costis, engineering manager of the Adversary Research Team at AttackIQ, said the warning should be taken seriously by critical infrastructure operators given traditional targeting methods by threat groups.

Recent attacks on critical infrastructure have often centered on “gaining persistent access, stealing sensitive information or positioning inside networks for future operations,” he noted.

"This latest exposure poses a bit of a different threat. The danger is more immediate because PLCs sit much closer to the physical processes that keep critical infrastructure running. Where a compromised business system can expose data, a compromised PLC can interfere with manufacturing, energy, water, or other physical processes. The downstream effects can reach far beyond the initial target."

The NSA stresses that the risks extend beyond the Siemens PLCs highlighted in the advisory, saying they represent just one subset of the wider threat landscape.

FOLLOW US ON SOCIAL MEDIA

Colt Data Centre Services (Colt DCS) has announced the appointment of Quy Nguyen as chief executive officer (CEO), effective immediately.

Nguyen steps into the role having served as acting CEO since April 2026, following the retirement of Niclas Sanfridsson.

Since first joining Colt DCS in 2016, he has held a number of senior leadership positions across sales and marketing, customer experience, design, and delivery business functions.

In his most recent role as chief sales officer (CSO), he played a key role in strengthening the firm’s commercial strategy and customer relationships.

Prior to joining the business, Nguyen also served in senior leadership roles spanning finance, strategy, and general management, where he built experience in scaling businesses and delivering growth.

As CEO, he will now lead Colt DCS through its next phase of expansion, as it looks to scale its global platform and support customers’ growing digital infrastructure requirements.

In an announcement, Colt DCS chairman Tim Cohen said Nguyen’s “exceptional leadership” has been a key factor in the company’s significant recent growth and transformation.

“His deep understanding of our customers, our people, and our business, combined with his strategic vision and proven track record of execution, made him the outstanding choice to lead the company,” he commented.

“As demand for digital infrastructure continues to accelerate globally, we are confident that Quy will build upon our strong foundations and position Colt DCS for continued long-term success."

Colt DCS operates 15 data centers globally, with a further 12 sites under development across nine cities in the UK, Europe, and the APAC region.

With nearly 800MW of capacity under development across its key markets, the company said it is well-positioned to support the growing demand for digital infrastructure that is being driven by continued cloud adoption and the rapid expansion of AI workloads across Europe and Asia.

"I am honoured to be appointed CEO of Colt DCS at such an exciting time for our company and industry,” Nguyen commented. "We have built a strong track record for delivering world-class digital infrastructure, fostering trusted customer relationships, and executing ambitious growth plans across key markets.”

Nguyen said Colt DCS will support customers’ evolving needs through continued investment, operational excellence, and innovation, as demand for sustainable cloud and AI infrastructure continues to grow.

“I look forward to working alongside our talented teams around the world to build on our strong foundations, expand our global platform, and deliver long-term value for our customers, partners, and stakeholders," he added.

FOLLOW US ON SOCIAL MEDIA

Cloud development company Vercel has dared cyber researchers to break out of its security sandbox environment as part of a $1 million challenge.

The two-week program will be run through HackerOne and challenges participants to break out of an isolated sandbox hosted by the company.

This is a two-pronged challenge, according to Vercel, requiring hackers to escape compute boundaries to reach an EC2 host, reach another tenant’s sandbox, or crash another tenant’s sandbox.

Network boundaries are also in the crosshairs, the company noted. Participants are challenged to “defeat the sandbox firewall” and reach unauthorized destinations, infiltrate data, or scoop up credentials.

Vercel said bounties will be paid per report, with a maximum pay-out of $50,000 for identifying a vulnerability that allows a threat actor to “read or modify another Vercel tenant’s data”.

Under the hood of the Vercel Sandbox

In a blog post detailing the challenge, Vercel said its sandbox environment runs on bare-metal EC2 hosts, with each sandbox given its own Firecracker microVM with a “dedicated guest kernel”.

“Inside that microVM a Linux container runs the operator’s code,” the company noted. “The microVM, not the container, is the security boundary, so operator-supplied code runs two layers removed from the host.”

Meanwhile, the network boundary is enforced on the host outside of the microVM.

“The sandbox firewall intercepts outbound TCP and DNS, checks each connection against the operator's domain and CIDR policies, and can inject credentials at the boundary so they never enter the microVM.”

Sandbox security in the spotlight

In a statement on X, CEO Guillermo Rauch said the challenge comes in direct response to high-profile incidents involving AI agents.

Agents at OpenAI, Anthropic, and Meta all breached containment during recent testing schemes, thereafter waging attacks against other companies such as Hugging Face.

As ITPro reported in early August, the sandbox testing environments used by all three companies were hosted by a third-party provider, Irregular.

“We are putting $1m towards verifying the security of Vercel Sandbox, in the open,” he said. I'm looking forward to bringing transparency to what frontier models can and cannot do in terms of real-world guardrail exploitability.”

In the event that participants do escape the Vercel Sandbox, Rauch said the company will be “ready to patch, iterate, and share our findings with the broader community”.

Vercel pointed to a recent test which saw the company’s CTO use an open-weight model to try and break sandbox containment. While the firm noted it did not escape, it “mapped the guest kernel, built a VM to reproduce its ideas, and wrote a fuzzer”.

FOLLOW US ON SOCIAL MEDIA

The number of victims affected in the CareCloud breach has been revised from 350,000 to roughly 3.75 million, the company has revealed.

The American medical record storage first confirmed it had been breached back in March, with hackers gaining access to medical data held in its cloud over six whole days.

According to regulatory filings from CareCloud at the time, the intrusion was spotted on 16 March. An investigation into the incident found an unauthorized party had gained access to an AWS environment for several days.

The initial filing said the attack caused a network outage of eight hours, disrupting access to the impacted database.

CareCloud noted the hackers "claimed to have exfiltrated data from databases within that environment", though the nature of the attack and whether CareCloud was directly communicating with the attackers remains unclear.

Reports suggest no hacking group has as yet claimed responsibility for the attack.

After CareCloud spotted the incident, access was shut down and no further unauthorized activity was spotted.

CareCloud breach notices cited far lower victim numbers

In July, disclosure letters were distributed to around 350,000 people in the US who were believed to be affected — but that has since risen to 3.75 million potential victims.

Leaked data includes names and addresses, as well as more sensitive details such as bank accounts, payment card numbers, medical data, and government identification including driver's licenses, passports, and Social Security numbers.

Ross Filipek, CISO at Corsica Technologies, said the revision to the victim list highlights the disastrous impact of cyber attacks on healthcare organizations – which rank among the top targets for cyber criminals.

"This isn't just a massive breach of data, it's a warning," Filipek commented. "Nearly four million patients having their information exposed shows just how much sensitive data can be concentrated behind a single healthcare technology provider."

"It also creates significant legal and regulatory exposure for a provider handling this much protected health information, which could make the fallout expensive for CareCloud."

What's next?

CareCloud hasn't revealed much detail about the attack – including whether it was ransomware or a ransom was paid – but the company says its continuing to investigate and lockdown systems.

"Upon discovering the incident, CareCloud quickly launched an investigation and took steps to contain and remediate the issue," a spokesperson said, per reports from Teiss.

"CareCloud engaged external cybersecurity experts and, with their assistance, secured the affected environment, eliminated the threat, and confirmed that no persistent unauthorized access remained. CareCloud is continuing to strengthen the security of its systems and environments."

ITPro approached CareCloud for comment, but did not receive a response by time of publication.

Disclosure letters distributed to affected patients note the company has seen no attempts at identity fraud to date, but warns recipients to be mindful of misuse of the stolen information.

Data exposed in the breach gives cyber criminals “plenty to work with long after the initial incident is over,” Filipek said.

Hackers often use exposed information such as email addresses and phone numbers in follow-up scams in the wake of a cyber attack or breach.

"For patients, the risk doesn’t stop at identity theft," noted Filipek. "Stolen health information can fuel highly convincing phishing, medical fraud, and other scams built around deeply personal details."

FOLLOW US ON SOCIAL MEDIA

NIST has issued a call for advice on how it can modernize the National Vulnerability Database (NVD) in light of recent AI advances – cyber experts have told ITPro that it’s desperately needed.

In a recent request for information (RFI), the institute said it is seeking stakeholder input on how to overhaul the vulnerability reporting service. The NVD plays a key role in helping organizations keep their finger on the pulse of the threat landscape.

The database acts as a catalog for software and hardware-related security flaws, drawing on the Common Vulnerabilities and Exposures (CVE) system to provide details and guidance on vulnerabilities.

Yet recent AI advances, particularly in the cybersecurity field, have raised the stakes when it comes to vulnerability identification and remediation.

Rob O’Connor, EMEA CISO at Insight, told ITPro that “the economics of vulnerability discovery have changed” due to AI. Humans simply can’t match the speed of agents, and that creates blind spots for security teams.

“Automated agents can analyse and triage code at a scale humans can no longer match,” he said. “This means the bottleneck now isn’t finding the vulnerabilities. Instead, it’s in contextualising and remediating them.”

Researchers at Forescout told ITPro earlier this year that businesses should prepare for an explosion of vulnerabilities, with AI now being used to identify software flaws at record pace.

Some big tech companies have already taken action on this front, including Apple. In late June, the company revealed plans to speed up software patching in direct response to AI.

At the core of NIST’s request, the institute said the goal is to “improve the NVD’s scalability, automation, interoperability, transparency, and utility”.

Put simply, AI is speeding things up, and the NVD needs to match the pace of vulnerability disclosures, which the institute itself recently admitted has become untenable.

In April, NIST revealed it would scrap efforts to analyze every submitted CVE, instead focusing on only the most severe vulnerabilities. The shake-up means that only CVEs that meet certain criteria will be ‘enriched’ with additional details and guidance.

The NVD was built around human speed

A key issue with the NVD in its current state is that it was designed around “human speed”, according to Crystal Morin, senior cybersecurity strategist at Sysdig.

The database launched in 2005, having evolved from an earlier setup known as the Internet-Categorization of Attacks Toolkit (ICAT). The scale – and indeed, speed – of threats at that time pales in comparison to what organizations face in 2026.

That’s not to suggest that reporting standards are obsolete, though - they just need a rethink to remain fit-for-purpose in an AI-powered era.

“They were designed for human speed. We now operate in a machine-speed world. A CVE with a static severity score tells you a flaw exists and, roughly, how bad it could be in theory. It doesn’t tell you whether it is exploitable in your environment, or if it’s already being weaponized,” she told ITPro.

With AI in the mix, Morin said this creates two distinct gaps: namely speed and the detail required to act on potential vulnerabilities.

“The reporting and enrichment pipeline still moves in weeks while exploitation moves in hours,” she said. “Second, the same detailed advisory that helps defenders is also raw material for AI-generated exploit code.”

“Ultimately, severity scoring on its own drives the ‘patch everything’ behavior that simply does not scale.”

Morin noted that Sysdig’s threat intelligence unit saw this dynamic play out in real-time when a Langflow vulnerability with a 9.9 CVSS score “essentially sat untouched while a lower-scored 9.3 vulnerability in the same product was mass exploited”.

“Standards must evolve toward data that is real-time, machine-readable, and grounded in real-world exploitability rather than theoretical severity.”

Designing for machine consumption

Efforts to modernize the NVD should focus primarily on designing it for “machine consumption”, according to Rob O’Connor, EMEA CISO at Insight.

With AI playing an increasing role in vulnerability management, curating easily consumed information could help speed up reaction times for security practitioners.

“I’d recommend designing it primarily for machine consumption, with human users as a secondary consideration,” he told ITPro. “As vulnerability management becomes more automated, data needs to be structured so machines can interpret and act on it easily, while remaining clear and useful for human analysts.”

Douglas McKee, director of vulnerability intelligence at Rapid7, noted that the ecosystem has already begun moving toward “structured enrichment” through authorized data publishers (ADPs).

These are organizations that are authorized to “enrich the content” of CVE records, a practice that NIST noted earlier this year that it will roll back.

Elsewhere, standards such as the Common Security Advisory Framework (CSAF) and the Vulnerability Exploitability eXchange (VEX) also provide machine-readable information on what specific products are affected by flaws.

“Those are exactly the kind of building blocks an automated vulnerability management system needs,” McKee told ITPro.

“I would move the NVD toward a federated enrichment model rather than trying to make NIST the place where every piece of vulnerability analysis has to happen. Let CNAs, vendors, researchers, and qualified data publishers contribute structured enrichment while the NVD acts as the trusted aggregation and normalization layer.”

Morin echoed McKee and O’Connor’s comments regarding machine-speed upgrades to the database.

“My core recommendation would be to evolve the NVD from telling defenders a vulnerability exists to telling them the extent to which it matters in production environments, in real time and in a form machine-speed security can act on,” she said.

FOLLOW US ON SOCIAL MEDIA

Fake North Korean IT workers are taking their job-hunt seriously, submitting as many as 60 job applications per day according to new research.

Groups of North Korean IT workers, dubbed PurpleDelta by Recorded Future, created at least 22 fabricated personas, with job applications being made across a range of recruitment websites and platforms such as LinkedIn and Upwork.

Notably, the group is using identity-brokering services, account-renting via AnyDesk, and multi-accounting tools.

Researchers found these fraudulent workers are often coordinating via Telegram and Slack, with support from facilitators who procure and maintain company-issued hardware on the operators' behalf.

“A successful job placement provides the PurpleDelta operation with a steady income and places a false employee within a company’s normal systems," said Alexander Leslie, senior advisor at Recorded Future.

"Wages are often funnelled toward sanctioned North Korean military and nuclear programs, and access may also expose information that was never meant to leave a company.”

Fake North Korean IT workers are turning to AI

The fake workers' applications typically include AI-generated profile photos, custom-configured ChatGPT assistants, and identity documents sourced from an illicit ID-generation service.

In terms of targets, researchers noted they’ve been applying for jobs at software and technology companies, as well as healthcare and biotechnology firms. Some appear to have been successful, according to Recorded Future, infiltrating at least 10 organizations.

Once inside, the fraudulent workers recorded internal meetings, used screen recording software during work sessions, and drafted pre-written Google Translate excuses to justify the use of personal devices and personal bank accounts.

“The scale of PurpleDelta’s operation is easy to miss when a company sees only one application," said Leslie.

"During interviews, PurpleDelta operators copied transcribed questions into ChatGPT and read the answers back, sometimes word for word, and occasionally repeated incorrect answers. A candidate can sound prepared without fully understanding what they are saying, which makes ordinary interview cues less reliable."

Manipulating company hardware

In some cases, Recorded Future said fraudulent workers received a company laptop and kept it and connected it in the country where they claim to live - usually the US, Germany or Brazil.

In one instance, researchers spotted one operative managing at least four identities simultaneously.

“PurpleDelta can recover quickly when one identity is exposed and a persona can be replaced," said Leslie.

"The same application machinery can keep running under a new name, and organizations should expect these operatives to adjust their methods as hiring teams become more familiar with them. Continued verification gives companies a better chance of catching changes."

Researchers warned that these fraudulent workers have been highly successful so far, collecting high-value intelligence and exfiltrating proprietary data, source code, and internal communications in support of North Korean state objectives.

"Companies should verify an employee’s identity after hiring, checking this during onboarding, so they can confirm where a worker and the hardware actually are," Leslie said.

FOLLOW US ON SOCIAL MEDIA

Fortinet has announced the acquisition of Virtue AI, an AI security specialist that provides runtime protection, automated validation, and security for autonomous AI systems.

The move will see Virtue AI’s capabilities added to Fortinet’s existing AI security portfolio, which includes FortiGate Hyperscale Firewall and FortiAIGate.

The cybersecurity giant said the deal will strengthen its ability to protect AI models, applications, and agentic systems across the AI lifecycle, from development through to runtime.

Founded in 2024, Virtue AI’s platform includes automated red-teaming of autonomous AI agents, agent protection and governance, continuous AI validation, and real-time security guardrails across text, images, video, audio, and AI-generated code.

Virtue AI's Guardian Agent technology provides visibility into AI agents and tools running within an environment, while also scanning Model Context Protocol (MCP) tools and source code for hidden security risks and monitoring agent behavior.

The San Francisco-based company’s automated validation capabilities are designed to identify new risks across model updates and policy fine-tuning, with testing covering hundreds of attack vectors and more than 1,000 risk categories.

Virtue AI’s red-teaming capabilities test autonomous agents across more than 50 sandboxed environments and 14 high-stakes domains, including simulated prompt injection and MCP-based attacks against leading agent frameworks.

Expanding protection across the AI lifecycle

Fortinet’s latest acquisition comes at a time when organizations are deploying more AI applications and autonomous agents, expanding the attack surface to include prompts, models, agents, MCP tools, APIs, and AI infrastructure.

According to Gartner, the market for securing AI ecosystems and AI agents is expected to grow from $2.8 billion in 2026 to $16.4 billion by 2030 – an expansion that Fortinet believes reflects growing demand for security technologies as AI adoption continues.

Earlier this year, the vendor introduced FortiAIGate in a bid to protect large language models (LLMs) from risks such as prompt injections, data leakage, model poisoning, and excessive resource consumption.

With the addition of Virtue AI, Fortinet said it is gaining technology that complements FortiAIGate by extending protection to AI models, applications, and agentic systems, while adding automated validation and real-time protection.

Commenting on the acquisition, Fortinet founder, chairman, and CEO Ken Xie said Virtue AI’s capabilities strengthen the company’s goal of providing continuous security for enterprise AI systems.

“AI is fundamentally changing enterprise computing, and security must evolve just as quickly,” he explained. “Virtue AI’s technology will advance our vision for continuous AI assurance, helping customers govern and protect AI systems throughout their lifecycle while operating them confidently at enterprise scale.”

Financial terms of the acquisition were not disclosed but Fortinet said the amount paid as consideration is “immaterial” to its business.

FOLLOW US ON SOCIAL MEDIA

AI in IT service management (ITSM) departments may be delivering a reasonable return on investment, but it's not yet making life any easier for IT professionals.

A new survey by SolarWinds found that 84% of respondents believe AI has met or exceeded ROI expectations, and many have unlocked meaningful time savings across core tasks.

Yet despite these apparent benefits, more than half (52%) revealed their overall workload has increased since adopting the technology. Notably, just 7% said the cost of AI adoption has matched what they planned for.

Even after an average of about 16 months using AI in ITSM environments, most teams are still managing AI’s overhead rather than realizing its full potential.

Real gains, but new workloads

SolarWinds said the survey highlights the productivity benefits of using the technology in ITSM tasks.

Respondents reported that AI is saving them an average of 3.2 hours per week detecting and flagging issues, 3.0 hours on end-user requests, and 2.9 hours on ticket triage.

Most of the time saved is being swallowed up by new work, however. Nearly half (48%) said they now spend time managing and maintaining AI tools and integrations.

A similar number (47%) spend more time reviewing and validating AI-generated outputs, while 37% end up training and fine-tuning AI models.

As for unexpected costs, 48% cited staff training, 47% data quality and clean-up, and 45% tuning and maintenance - all of which are ongoing, rather than one-time costs.

More than four-in-five respondents (83%) said they now spend three or more hours per week just keeping their AI systems running reliably.

ITSM teams still stuck in reactive mode

Most ITSM teams seem to be taking a reactive rather than proactive approach. According to SolarWinds.

When asked where AI has had the greatest impact across the incident lifecycle, three-in-ten cited identifying issues before they impact users and 23% pointed to prioritizing and routing issues.

Only 19% cited preventing issues before they occur as the area of greatest impact.

The answer, according to SolarWinds, is to concentrate efforts where they're most likely to pay off – high-frequency, well-defined tasks where gains are measurable and feedback loops are tight, such as ticket triage, issue detection, and incident.

Teams should consolidate AI closer to existing service workflows rather than spreading it across disconnected tools and strengthen the data foundation.

Notably, data quality is the top reason AI fails to deliver expected value, and treating it as part of the AI strategy rather than a separate clean-up project directly determines output quality.

“We’re at an inflection point in IT service management. AI adoption is no longer the hard part — the hard part is building the organizational discipline to make AI actually deliver,” said Brad McGinity, GM of ITSM, SolarWinds.

“The teams that get this right aren’t just running a faster service desk; they’re running a fundamentally different operation.”

Lingering ROI woes

ROI with AI has become a common recurring talking point for IT leaders over the last three years, with the financial benefits still up for debate at some enterprises.

Recent IDC research found one-in-five firms admitting they're investing aggressively in AI with little evaluation of the likely ROI.

However, ROI isn't necessarily all about the money. According to recent research from KPMG, other key metrics include the performance and quality of work and the speed and accuracy of decision making.

FOLLOW US ON SOCIAL MEDIA

Vibe coding startup Cursor has launched a new code hosting platform in a move that could put it in direct competition with GitHub.

Dubbed Origin, Cursor said the new service is “starting with the essentials”, offering users all the staple features found in GitHub. This includes code repositories (repos) as well as the ability to browse and edit codebases and handle pull requests.

Origin can also be synchronized with GitHub, the company revealed, meaning users can mix and match with both platforms.

“Your GitHub repos can sit alongside the ones Cursor hosts,” the company explained in a blog post. “Connect GitHub to Cursor, pick your org, and you'll see the repos you can sync. Select one and Cursor pulls it in.”

“You choose what gets synced and can disconnect a repo at any time. Anyone with read or write access to a synced repo can view it in Cursor too.”

Access to Origin has already begun rolling out for all paid plans, according to Cursor, albeit in an early beta capacity.

Long-term, Cursor hinted that the plan is to provide “agent-native features” akin to what it already offers through its agentic coding service.

The startup is also in the process of developing a broader “app ecosystem” to integrate third-party tools within Origin. Integrations with Vercel, Depot, and Buildkite are already available, with “more coming soon”.

“Connect Vercel from a repo's Apps tab and every PR gets a preview deployment where you can test and make comments,” the company explained.

“Merge, and it ships to production. For CI, connect Depot or Buildkite. Both run your existing GitHub Actions workflows and Buildkite also runs its native pipelines.”

A blossoming GitHub challenger?

Origin could represent a major challenge to GitHub moving forward, and as ITPro reported earlier this year, the Microsoft-owned company has been making moves to counter competition on this front.

Reports in January revealed Microsoft had reshuffled internal teams to stave off competition from challengers such as Anthropic and Cursor. First reported by Business Insider, this strategy focused primarily on expanding agentic AI coding options for users.

Sources told the publication that Microsoft had grown concerned about up-and-coming startups muscling in on GitHub’s market position.

These concerns centered largely on agentic coding capabilities, however. Origin muscling in on GitHub’s bread and butter as an all-encompassing developer platform marks an entirely different challenge.

Cursor faces significant hurdles if it is to mount a challenge though. Figures published by GitHub last year suggest the platform boasts over 180 million active users, and the platform continues to grow rapidly.

GitHub frustration

The Origin announcement also comes at a tricky time for GitHub. The developer platform recently experienced a lengthy outage that saw developers unable to access key services – and that marked the latest in a string of outages.

After a series of incidents this year, GitHub detailed plans to improve capacity amidst grumbling user concerns.

In an April blog post, GitHub CTO Vlad Fedorov attributed some outages to the sheer volume of users now operating on the platform, which is pushing infrastructure to the limit.

In October 2025, the company began work to boost capacity by ten-times before revising this in February 2026 with plans for a 30-fold increase.

FOLLOW US ON SOCIAL MEDIA

Wiz claims its AI found a flaw written and reviewed by GitHub Copilot, but the developer platform has pinned the blame on boring old human error.

In a blog post this week Wiz Research said that its Red Agent, an AI-powered bug hunting tool, had spotted a vulnerability in one of Snowflake's public repositories.

That is very much what Red Agent is designed to do, but the Google-owned security company claimed that the flaw had actually been introduced by AI itself, pinning the fault on GitHub Copilot.

"This incident highlights a new reality in software development: Critical vulnerabilities can still be introduced and approved within workflows involving AI coding agents and can still pass established automated security checks," wrote Gal Nagli, head of offensive security at Wiz..

Except that's not exactly what happened here. Wiz researchers noted that the flaw listed “Copilot Autofix powered by AI” as a co-author. After sifting through the commits, The Hacker News noted that the section with the flaw was actually attributable to a Snowflake engineer, while Copilot changed certain aspects.

Wiz has since updated its blog post to clarify this. In a statement given to ITPro, Wiz co-founder and CTO Ami Luttwak offered additional details on the incident.

“The relevant PR was co-authored by multiple contributors including Copilot. Initially the blog implied the vulnerable code flow was generated by AI,” Luttwak explained.

“Soon after the blog was published, we issued an update following community feedback to clarify that the specific lines of code that caused the vulnerability, were not created by copilot (although it is mentioned as a co-author). Note that copilot did participate in the PR and contributed code and also scanned these specific lines via GitHub advanced security (AI code scanner) that missed the vulnerable flow.”

Luttwak noted that the case shows that clear attribution between humans and AI is “becoming a bit harder to establish”.

“Just looking at co-authors of the PR is not enough,” he added.

Wiz did find a Snowflake flaw

Regardless of the co-author confusion, a flaw in Snowflake was indeed found.

Wiz Red Agent spotted a script injection vulnerability that would have allowed an unauthenticated user to execute arbitrary commands “within a GitHub Actions runner by opening a GitHub issue with a specially crafted title," the company noted.

Even if Copilot didn't write the flaw, Wiz said the issue still wasn’t spotted when it was reviewed.

"GitHub Advanced Security scan analyzed the final PR revision, including the vulnerable workflow, but did not flag the critical injection," Nagli added.

The flaw was responsibly disclosed at the end of June to Snowflake, which quickly fixed the issue and confirmed that no-one but Wiz had made use of the vulnerability.

"Snowflake appreciates Wiz's responsible reporting of and collaboration around these findings through our vulnerability disclosure and bug bounty program, HackerOne," Snowflake said via a statement on Wiz's blog post.

"The disclosure was received on June 23, 2026, and it was immediately investigated and remediated, and our investigation found no evidence of unauthorized access."

Indeed, the flaw had only been live five days before Wiz Red Agent spotted it — and perhaps that's the real story, from published to uncovered to fixed in just five days, it's clear AI is speeding up the pace of security.

"The vulnerability was live for only five days before an automated agent discovered and validated it," Nagli said in the blog post. "Security operations must adapt to a landscape where automated discovery occurs in hours, requiring rapid patch cycles and short-lived credentials."

FOLLOW US ON SOCIAL MEDIA

Snowflake has announced new dynamic model routing capabilities for flagship products along with expanded access to open-weight models for customers.

The new capabilities build on Cortex AI Gateway, according to the firm. It was unveiled in July this year and provides customers with a “unified foundation” for agent governance and routing requests.

The aim here, the company noted, is primarily focused on optimizing AI token consumption, which has become a recurring pain point for enterprises over the last 12 months.

Snowflake said the new capabilities mean Cortex AI Gateway will “automatically select" a specific model based on the quality and cost for particular tasks.

Put simply, this will see Snowflake assign more efficient models for lower-complexity or repetitive tasks. Meanwhile, tasks that require “deeper reasoning” are routed to frontier models.

This means customers can reduce unnecessary inference spend and remove the need for laborious manual model selection for each individual task, according to the firm.

Snowflake CEO Sridhar Ramaswamy said the update comes in direct response to growing concerns over AI-related costs. Token costs have surged so far across 2026, while a host of major providers have switched to new consumption-based models, resulting in increased costs for enterprises.

“Enterprises are becoming much more rigorous about the economics of AI. The question is no longer how much AI they are using, but whether that AI is translating into meaningful business value,” said Ramaswamy.

“Achieving intelligence efficiency requires the flexibility to use the best model for each task as the landscape evolves. Snowflake’s role is to absorb that complexity so customers can focus on outcomes while we optimize model choice underneath.”

Dynamic routing will also be integrated across Snowflake CoCo and Snowflake CoWork, and will be made available for third-party AI agents.

Snowflake expands model library with open-weight options

In addition to model routing capabilities, Snowflake revealed it will expand access to several leading open models, such as DeepSeek-V4-Flash 0731 and GLM-5.3.

The addition of these models adds to what is already an extensive library available for customers, Snowflake said. Enterprises can already choose from a variety of models from Anthropic, Google, Meta, Mistral, and OpenAI.

“This portfolio gives customers more freedom to choose the right combination of performance and cost for each workload,” the company said in a blog post.

Once again, the focus here is firmly on helping to manage costs by providing a broader range of options.

Internal testing conducted by the firm found that enterprises using a combination of open and proprietary models can “deliver comparable quality while materially improving efficiency”.

In one evaluation detailed by the company, agents using dynamic model routing through Cortex AI Gateway were able to build a data build tool (dbt) pipeline with up to three times greater token efficiency compared to a frontier model-only approach – all while maintaining the same quality.

“In a separate test, engineering teams completed the same number of pull requests with 25 percent greater token efficiency,” the company said in a blog post.

FOLLOW US ON SOCIAL MEDIA

Spirit Airlines collapsed in May this year, but its data is set for a new lease of life after Google acquired it in an auction.

The tech giant recently outbid AI company Mercor to acquire the de-identified data, according to court documents. Google bid $10 million, versus Mercor's $7.5 million, though reports note the bidding started around the $5 million mark.

A third bidder also took part but was not named.

A spokesperson for Google told ITPro the aim behind acquiring the defunct airline’s data lies in its potential use for AI training and product refinement.

"We acquired part of an enterprise dataset from Spirit Airlines, which can be helpful in improving our products and AI models,” the spokesperson said.

That same sentiment was highlighted by a spokesperson for Mercor, per reports from Business Insider.

"Companies are sitting on decades of records that show how real work gets done, and that data is now some of the most valuable material for training and evaluating AI," the spokesperson said.

"We partner with leading companies to license their operational data to the labs building the next generation of models. Spirit was that same process applied to a bankruptcy estate."

What Google bought

The data bought by Google includes all productivity and collaboration data, which the filing notes includes everything from emails to chats, spreadsheets to calendars, as well as "core business systems and business applications data".

This includes data on employee behavior and productivity, aircraft operations, and inventory, among other topics. Beyond that, Google has purchased workflow and process data from departments such as HR and marketing.

In total, Google is picking up 100 million emails, 500 million Microsoft Teams chats, 30 million lines of code – but not records of loyalty program members or other customer data.

Indeed, the data set does not include any consumer data or personal data, the filing noted, or any other information that could fall under data protection laws. The buyer was required to agree to not attempt to re-identify the data and link it to any person.

The buyer is allowed to transfer the data set to third parties, however.

"We will not receive any personal information from this dataset," the Google spokesperson added. "Any data we receive will be rigorously scrubbed of any personally identifiable information by a third party before receipt."

Alongside the data, Google's $10 million also buys it all of Spirit's internally developed software and applications, including all the source code, plugins, data files, libraries, APIs and documentation.

The battle for data

The battle for the business data reveals how desperate AI developers are for fresh material to feed AI, in particular for training models, though Google didn't specifically say that was the purpose of the data set.

Anthropic has raised eyebrows via a campaign of buying up pre-2022 books to scan for training data, while Google was reportedly trying to buy code from Google Play Store app developers.

SimpleClosure helps failed startups sell their old code, chats, and emails to AI companies, and Meta had to pause plans to track its own employees for AI training purposes amid internal backlash.

All of this is to avoid issues like data exhaustion or hitting a data wall, in which they lack enough good quality data to build bigger models, and to avoid model collapse, when training data isn't of good enough quality — or is full of AI generated content.

FOLLOW US ON SOCIAL MEDIA

A hacker is advertising data allegedly stolen from an array of companies including McDonald's, Vodafone, Kyndryl, and more.

Researchers at HudsonRock said that the criminal, known as TheHatman, downloaded the databases directly from the organizations’ Azure or Entra portals, claiming to have done so through the use of compromised credentials.

Other victims include HCL Technologies, IHG Hotels & Resorts, Gap, Hexaware Technologies, and Wyndham Hotels & Resorts.

"The campaign impacts multiple global enterprises across IT services, hospitality, telecommunications, retail, and logistics," researchers said.

"Our researchers went over the data and it seems highly legitimate based on the corporate email addresses found, combined with field names perfectly matching standard Azure directory exports."

The data – which includes roughly 3.6 million records in total – appears to include full names, corporate email addresses, including active domains and tenant-specific .onmicrosoft.com structures, phone numbers, and physical addresses.

It also covers employee IDs, job titles, departments, notes, manager details, and direct reports, along with user group memberships, service accounts, and highly privileged account records such as Global Administrator listings.

"The exposure of service accounts and global admin names is particularly concerning, as this provides a direct roadmap for subsequent social engineering, spear-phishing, or targeted privilege escalation attacks against these organizations," HudsonRock said.

TheHatman's methods still unclear

Exact details on how the threat actor gained access to this trove of data remains unclear, according to researchers.

Possibilities include infostealer malware infections which compromised employee session tokens, phishing campaigns, or a lack of strict multi-factor authentication (MFA) on specific tenant portals.

Researchers also suggested it could be down to abuse of a third-party API that had excessive read privileges across multiple environments.

According to HudsonRock, the most likely explanation is a targeted exploitation of infostealer infections rather than a systemic zero-day vulnerability in Azure, as this would have hit a far broader range of victims.

Alleged victims react

Companies said to have been caught up in the breach have begun taking action.

TCS, for example, has filed a statement with India’s stock exchange, revealing that it hasn’t yet found any credible evidence of a breach of systems or customer environments.

"The information referenced appears to be more than four years old and limited to basic employee information. There is no indication that customer data, customer systems, or TCS operational systems have been impacted," the statement reads.

ITPro approached the other aforementioned victims, but did not receive a response by time of publication.

Valuable credentials

Darren Williams, CEO and founder of BlackFog, said that if the claims are legitimate then this once again highlights the value of exposed credentials for cyber criminals.

These can often represent the keys to the castle for hackers, giving them sweeping access to enterprise environments.

“The reported use of leaked credentials to access Azure and Entra environments shows how valuable compromised identities have become to cyber criminals," he said.

"MFA, strong identity controls and employee awareness remain essential, but organizations must assume credentials can be compromised. These measures must be backed by technology that prevents sensitive data from being exfiltrated, even when attackers successfully gain access to the network.”

UPDATE

A spokesperson for Vodafone told ITPro: “We have been investigating a claim that has been made in relation to Vodafone data. Our current assessment is that data is old Vodafone employee information that would be available on a business card. The information available could be accessed from an address list.

"No Vodafone customer data or impact has been identified. We would like to assure our customers that we take security very seriously. Our security tools are constantly updated and we proactively track new threats and vulnerabilities. We have an international team of cyber security professionals who constantly protect, defend and monitor our systems.”

A spokesperson for Gap told ITPro: “Our preliminary investigation indicates that the data in question is limited in scope, non-sensitive and dated back to several years ago. Notably, there is no evidence to suggest that our corporate systems have been compromised.”

FOLLOW US ON SOCIAL MEDIA

With data centers dramatically increasing pressure on the electricity network, the UK's National Grid has announced plans to improve power provision in North West London.

Five new data centers – equivalent to 1GW of demand – will be connected in Hertfordshire to the existing 400kV Elstree GIS substation, with a new 132kV Letchmore Heath GIS substation planned for next door.

Also in the pipeline are upgrades to Sundon substation in Bedfordshire, the reconductoring the 35km Sundon–Elstree overhead line route, and an extension to the existing Elstree gas insulated switchgear (GIS) substation.

Meanwhile, work on the route between Sundon and Elstree will see over 200km of overhead line upgraded, with 60km of new cable infrastructure also to be installed within the existing Elstree–St John's Wood tunnel.

Additional works include:

• The installation of a new 400kV cable circuit in the existing Elstree–St John’s Wood tunnel
• Upgrades to the Elstree–St John’s Wood tunnel and headhouses
• The reinforcement of the existing St John's Wood substation

“The North West London Upgrade is a nationally significant investment that will strengthen the electricity network, unlock new connections, support regional economic growth and help keep long term energy costs down for consumers," said Joe Senior, project director at National Grid.

“Working closely with our delivery partners and supply chain, we're bringing together substations, cables and tunnels, and overhead line reinforcement in a single program to deliver the infrastructure needed to power future homes, businesses and digital industries as electricity demand rises.”

In line with government guidance, communities near the new Letchmore Heath substation are set for £530,000 in funding for projects that improve the social, economic, and environmental wellbeing of the community.

National Grid eyes capacity gains

According to National Grid, these upgrades will increase network capacity and strengthen the resilience of electricity supplies across the region, helping to keep energy bills down.

It should also play a role in supporting economic growth across North West London as electricity demand rises – enabling new data center connections without jeopardizing power supply to homes and businesses.

As part of these efforts, National Grid is carrying out other ongoing work across the capital, including London Power Tunnels 2 (LPT2) and the North London Reinforcement (NLR) scheme.

Earlier this year, National Grid revealed that the queue for connections to the transmission network had grown by 460%, in large part because speculative applications were swamping the pipeline and delaying connections for strategically important projects.

The result, according to the National Grid, is that some waiting times have surged to up to 15 years.

Because of this, the government announced plans to prioritize AI data centers, AI Growth Zones, and industrial sites when it comes to new connections.

FOLLOW US ON SOCIAL MEDIA

A lengthy GitHub outage that saw users struggle to access core services has been resolved, according to Microsoft, but what happened?

The incident first came to GitHub’s attention at around 1:40pm UTC yesterday, when the company said it was investigating reports of impacted performance for some services.

It proceeded to report problems with API Requests, Actions, Webhooks, Pull Requests, Issues, and GitHub Copilot.

"We are experiencing high error rates around 20% for web experiences and api traffic. Archive downloads and raw repository content downloads are experiencing an approximate 50% error rate," the company reported about an hour into the affair.

"SAML and OIDC authentication, SCIM, and Team Sync are also impacted. Investigations are on-going and we will continue to provide updates as we discover more information."

By 4.40pm, the firm said it had identified the problematic component and had taken corrective action – although error rates were still slightly up.

“We are still seeing residual impact in the form of sporadic authentication failures. We are continuing to apply additional mitigations and investigate the remaining impact,” GitHub said.

Full GitHub outage analysis pending

At one point, Downdetector was reporting nearly 3,000 outage reports, although the incident was finally resolved by 9.15pm.

Exact details on the cause of the incident are yet to be revealed. However, GitHub promised that a “detailed root cause analysis will be shared as soon as it is available”.

GitHub demand surging

GitHub is one of the world's largest software development platforms, with 225 million users, but has seen its infrastructure pushed to the limit by the boom in AI-driven coding.

Last October, it started work to boost capacity by ten times before realizing in February that a 30-fold increase would be required.

In an April blog post detailing the strain placed on the firm, CTO Vlad Fedorov said agentic workflows have "accelerated sharply” in recent months.

“By nearly every measure, the direction is already clear: repository creation, pull request activity, API usage, automation, and large-repository workloads are all growing quickly," Fedorov wrote.

He said GitHub planned to deal with availability first, then capacity, and then new features.

"We are reducing unnecessary work, improving caching, isolating critical services, removing single points of failure, and moving performance-sensitive paths into systems designed for these workloads," Fedorov said.

Repeated outages

Notably, it appears the company wasn't able to carry out this work fast enough, with eight further outages last month. This included a lengthy GitHub Actions outage that the company described as 'unacceptable'.

The company attributed this to the fact that the GitHub Actions service at the core of the incident was still fully running in its data centers, and promised to accelerate its move of GitHub Actions to Azure.

Here, the company noted, it will have more headroom and capabilities to absorb sharp spikes in activity.

"GitHub is on track to see a 1,400% increase in commit volume from 2025 to 2026, with most of the increase coming from AI tooling," commented Jamie Beckland, chief product officer at APIContext.

"Every technology team depends on platforms that are being reshaped by new demand from AI. Which means every technology team now has an AI infrastructure problem. Technology teams need independent monitoring of the third-party services sitting in their delivery path so they know whether problems affect developers, pipelines, and customers."

FOLLOW US ON SOCIAL MEDIA

New figures from Gartner point to a surge in AI-optimized infrastructure spending over the next year. A senior analyst at the firm told ITPro it underlines how it is driving a “fundamental shift” in cloud consumption and design.

Recent analysis from the firm found spending on this front is projected to grow 96% across 2026, reaching a total of $42 billion as enterprises look to overhaul infrastructure with the specific aim of dealing with cumbersome AI workloads.

Hardeep Singh, senior principal research analyst at Gartner, told ITPro the consultancy expects this trend to continue as adoption rates rise. Indeed, Gartner projects the market to “sustain high growth” and reach $66 billion in spending.

“There is a fundamental shift in how cloud infrastructure is being designed and consumed,” Singh told ITPro. “We are transforming from traditional general purpose cloud computing towards purpose-built AI infrastructure designed specifically for large language models (LLMs) and agentic workloads”.

A key factor behind the optimized infrastructure push lies in the changing shape of AI use, Singh noted. The technology is maturing and operationalization of AI, particularly agentic AI, is “accelerating inference demand” across the board.

Agentic AI has a far larger token consumption rate compared to traditional chatbot-based AI interactions, which in turn increases inference demand.

Research from Signal65, for example, found agentic workloads consume “orders of magnitude more tokens”, with consumption rates anywhere between four to 15-times more.

“You see [with] agentic AI it has a variable consumption,” Singh told ITPro. “It would be one prompt triggering autonomous model calls, and that would lead to a huge amount of compute consumption.”

Gartner projects that inference demands will surge over the next year on the back of increased agentic AI adoption. Indeed, the consultancy noted that spending on this front will reach $23.3 billion, surpassing the $19 billion spent on AI inference.

The fact that inference spending will exceed training for the first time is noteworthy, Singh told ITPro. Many enterprises are still in the early stages of agentic AI adoption though. As projects mature, this is certainly expected to drive inference demands even further.

Gartner ranks AI use based on a level system, with basic chatbots at the first level, and fully autonomous at level five, the highest rating. Most businesses aren’t even close to that level, he said.

“It will still take some time to go and reach the autonomous side,” Singh told ITPro.

“We are still in that L1, L2, L3 kind of scenario where experimentations are happening, trying to understand which use cases fit in which scenario, and then how the agents [are] going to solve the complex business problems.”

Given the current state of agentic adoption and projected advances on this front, AI-optimized infrastructure will be even critical for meeting demand.

Gartner figures show that 55% of spending on this front will directly support inference. This is set to reach 59% by 2027.

Changing priorities

According to Gartner, the growing volume of inference workloads will ultimately “reshape cloud investment priorities”. Singh noted that this can already be observed in cloud strategies across a range of industries.

Research from IDC in late 2025 predicted a sharpened focus on hybrid and even private cloud models due to a combination of factors, most notably “more predictable cost models”.

Singh told ITPro he expects this focus to continue as agentic adoption accelerates, particularly with the rise of multi-agent frameworks in which they rely on several separate underlying AI models.

“The important thing will be how enterprises want to use multi-agents backed by multiple models, so multimodality and multi-models,” he said.

“Both these things are important in the context where some of the data workloads [are] moving to the cloud, training the large language model, then deploying on the edge on prem to get the inference and real-time monitoring.”

“Hybrid and multi-cloud will be more picked up in these scenarios, because some enterprises know that it is difficult to train the model on-prem because of the amount of capacity it needs.”

FOLLOW US ON SOCIAL MEDIA

HM Revenue & Customs (HMRC) has announced a deal with Atos aimed at driving IT modernization with low-code solutions.

The three-year deal, worth more than £78 million, will see Atos act as lead partner for Low-Code Leadership and Configuration Services within the Digital and Legacy Application Services (DALAS) framework.

In a statement announcing the move, Atos said it will provide “technical and delivery governance and assurance across the entirety of HMRC’s low-code ecosystem”.

This includes supporting the tax office’s IT modernization efforts and improving digital services for workers, businesses, and citizens across the UK.

“We’re proud to be expanding our long-standing relationship with HMRC through this appointment,” said Michael Herron, Head of Atos UK&I.

“Our proven technical expertise and experience in public sector transformation makes Atos well-placed to support HMRC as it looks to deliver better digital experiences for citizens at pace.”

HMRC targets low-code gains

Low-code tools allow businesses to accelerate transformation efforts by simplifying application development and removing manual programming.

According to Atos, the deal with HMRC will allow the tax office to ramp up legacy IT modernization and deliver “faster time to value” and cost savings.

The new contract builds on long-standing ties between Atos and HMRC. The tax office awarded a multi-year contract to the firm in 2022 to provide application testing services, for example.

It also represents the latest in a string of modernization-related announcements by HMRC in recent months. As ITPro reported in June, HMRC signed a deal with Capgemini, NiCE, and Route 101 to overhaul customer service operations with cloud native and AI tools.

The contract will see legacy infrastructure and existing products consolidated within a unified cloud native platform, with development led by Capgemini.

That deal came hot on the heels of a £175 million contract with Quantexa to overhaul its core data infrastructure.

Low-code tools still alive and kicking

Low-code platforms have been a staple in enterprise digital transformation programs for more than a decade.

Despite the advent of generative AI and an influx of AI coding tools in recent years, research shows low-code and no-code solutions are still going strong.

Research from App Builder in April 2025 found these tools still play a key role in enterprise digital transformation efforts worldwide.

Around 95% of companies revealed they still use low-code tools in software development. Notably, more than three-quarters (76%) said AI is helping make existing tools more efficient.

FOLLOW US ON SOCIAL MEDIA

Rackspace has announced the appointment of Chetan Gupta as chief AI officer, as the enterprise infrastructure specialist looks to advance its enterprise and sovereign AI strategy.

Gupta will lead Rackspace’s Office of AI, overseeing AI strategy, research, innovation, governance, and adoption across the company and its customers.

He joins the business following almost a decade at Hitachi, where he served in a number of senior AI leadership roles – including general manager of the Advanced AI Center in Japan, vice president of the Industrial AI Lab in North America, and head of Hitachi’s Global AI Center of Excellence.

Earlier in his career, Gupta also spent seven years at HP Labs, where he worked on deploying AI solutions across areas such as logistics, manufacturing, energy, and mobility. He has also authored nearly 300 papers and patents, as well as mentored AI researchers and engineers.

In an announcement, Rackspace CEO Gajen Kandiah said Gupta’s experience developing AI systems for critical environments will be a valuable asset as the firm advances its AI strategy.

"Chetan has spent his career building AI for the institutions the world depends on,” he explained. “At Hitachi, that meant trains, grids, and factories, where AI is not a demo, but an operating commitment measured in uptime, safety, and trust."

Kandiah said that experience will be particularly relevant as Rackspace customers move AI into production within regulated and sovereign environments.

"Chetan understands that success depends on more than model capability,” he added. “It requires strategy, governance, and an operator who carries accountability from silicon to outcomes.”

Rackspace eyes sovereign AI gains

Rackspace positions itself as an operator of the full enterprise AI stack, spanning governed private cloud through to AI inference and agents in production.

Its offering is built around an "Outcomes as a Service" model covering infrastructure, data foundations, and forward-deployed engineering.

The company said Gupta’s appointment will strengthen its ability to help organizations deploy AI safely, economically, and at scale – particularly across regulated industries, sovereign jurisdictions, and mission-critical operations.

Commenting on his new role, Gupta said AI has reached “an inflection point” as organizations shift their focus away from the technology’s capabilities and towards tackling deployment challenges.

“The hard problem is no longer capability. It is trust: taking promising research and turning it into systems that hold up in the real world, under real regulatory and operational constraints," he explained.

"Rackspace occupies a distinct position in the market, operating the full stack for institutions that cannot afford to get AI wrong. I look forward to leading our AI strategy and advancing enterprise and sovereign AI for our customers and across the business."

FOLLOW US ON SOCIAL MEDIA

Organizations are mostly seeing a return on their AI investment, new research shows, but many are struggling to translate business context into AI systems and workflows.

Eight-in-ten IT leaders globally expect AI spending to increase over the next two years, according to Alteryx's 2026 IT Leader Research: The State of AI Ownership, Agents, and ROI report.

While 77% agree business context - the rules, definitions, and operational knowledge that shape how their organizations operate - is critical to producing accurate and relevant AI outputs, 53% say they struggle to incorporate this within AI systems and workflows.

The result here is that AI models are forced to make generic assumptions that could lead to mistakes, false insights, or incorrect metrics – and that has a big impact on business efficiency.

More than a third of respondents told researchers that the ability to measure AI ROI will be one of the capabilities that most distinguishes technology leaders from their peers.

Notably, firms are increasingly measuring AI success through productivity improvements (53%), cost reduction (45%), and revenue growth or broader business impact (39%).

“Our research highlights a growing gap between AI ambition and enterprise-scale execution,” said Andy MacMillan, CEO of Alteryx.

"Organizations have proven they're willing to invest in AI, and many are already seeing returns. But scaling AI requires more than better models. It requires making the business knowledge people use every day available to the systems making decisions.”

Lacking context

While AI can analyze information and generate responses, it can't consistently apply company-specific rules, policies, thresholds, and decision criteria unless that knowledge is built into the workflows it uses to make decisions.

Part of the problem is limited data access, Alteryx found, with only 18% of organizations reporting that business users have fully self-service access to cloud data.

Indeed, most are still relying on IT or data teams for routine data access and analytics, with 38% describing a mixed model and 15% saying business users remain largely dependent on technical teams.

Two-thirds of technology leaders say AI and agent-based systems are most productive when managed within the line of business, with 71% saying that AI initiatives are most successful when IT and business teams collaborate closely.

However, strategy (37%) and delivery (38%) remain concentrated within IT, while business teams are most often responsible for defining requirements (30%).

"The organizations creating lasting value from AI will be the ones that operationalize their business logic so it becomes visible, governed, repeatable, and ready for AI," said MacMillan.

What’s driving AI ROI?

The AI investments that appear to be delivering the best ROI are workflow automation and autonomous agents, cited by 27%, followed by copilots or assistants, at 16%, and AI-powered customer experience at 14%.

Virtually all IT leaders (93%) told Alteryx they were confident agentic AI could deliver measurable ROI for their enterprise within the next two years.

The first processes to be automated by agentic AI, they reckon, will be IT operations and incident management, cited by 47%, followed by customer support and service workflows at 39% and data analysis and reporting at 36%.

"Within our research, a small group of organizations self-identified as leading the way in AI innovation, confirming a clear path to achieving meaningful business value with AI technologies," the researchers concluded.

"These organizations point to a clear set of priorities behind their progress: rigorous measurement of AI's business impact, treating AI rollout as a strategic and operational priority, and building the governance and literacy needed to scale with confidence."

FOLLOW US ON SOCIAL MEDIA

Dynatrace has agreed to acquire observability specialist Arize for $915 million, as the vendor looks to expand its capabilities across the AI development lifecycle.

The deal will bring together Arize’s AI evaluation technology with Dynatrace’s production monitoring capabilities, enabling customers to evaluate, operate, and continuously improve AI applications.

AI observability has become increasingly important as businesses move AI applications into production, connecting model and agent behaviour with application performance, GPU utilization, infrastructure health, and business processes.

Following closure of the acquisition, Dynatrace said customers will gain continuous coverage across the AI lifecycle, spanning experimentation and deployment readiness, through to runtime evaluation and observability.

The combined offering will also provide context of AI behavior and business impact, as well as an enterprise data foundation for AI workloads powered by exabyte-scale analysis and AI lakehouse capabilities.

The acquisition is expected to close later this quarter or early in Dynatrace’s third quarter, subject to regulatory review and customary closing conditions.

In an announcement, Dynatrace CEO Rick McConnell described the AI observability market opportunity as “enormous” with the category projected to exceed $10 billion by 2030.

“Acquiring Arize advances our AI observability leadership, accelerates our roadmap, enhances our long-term growth profile, expands our reach with the developer community, and adds an incredible AI-first team to Dynatrace,” he said.

Combining AI evaluation and observability

Founded in 2020, Arize provides AI observability and LLM evaluation technology designed to help development teams identify issues with AI applications, assess output quality, and monitor AI behavior.

The firm's platform is used across AI frameworks and model providers, with Dynatrace highlighting its open source community and developer-focused presence as part of the rationale for the acquisition.

The vendor said Arize’s addition will address fragmentation in the way AI applications are developed and operated, with AI engineering teams often using one set of tools to evaluate model and agent behavior, while another is used to monitor the applications and infrastructure beneath them.

Instead, the combined platform intends to connect AI evaluation and production observability to give developers and engineering teams greater visibility into issues across the AI application stack.

For developers, the deal also aims to provide a clear path from AI experimentation through to enterprise deployment, backed by Dynatrace’s existing production monitoring and observability capabilities.

Leadership continuity

Post-acquisition, Arize CEO and co-founder Jason Lopatecki, and fellow co-founder Aparna Dhinakaran, will both join Dynatrace. Lopatecki will continue leading the Arize team, reporting to McConnell.

Commenting on the acquisition, Lopatecki said the move will strengthen Arize’s focus on helping AI teams assess whether their agents are working correctly rather than simply running.

“Together, we can bring AI evaluation and software observability into an end-to-end system, enabling teams to build more ambitious AI applications faster,” he explained.

“This is the kind of innovation that only happens when two companies with highly complementary solutions and go-to-market models come together, setting a new bar for what AI Observability can deliver for the entire industry.”

FOLLOW US ON SOCIAL MEDIA

Cyber criminals are spending millions buying up expired domains to repurpose them for malware, scams, illegal streaming, and online gambling, according to new research.

Taking advantage of the domains' trust, backlinks, and web traffic, they're not only being used by the original threat actors, but are also being sold on.

Domains originally registered years ago, dropped, and then acquired by someone else can still carry signals associated with their history. Researchers, security products, and reputation-based algorithms may view them more favorably than a genuinely brand-new registration.

This makes these re-registered 'dropcatch' domains a hot property for threat actors. Analysis conducted by Infoblox Threat Intel observed around 65,000 being registered per day during the first half of this year, representing nearly one-in-five of all newly observed domains.

"The sheer volume of dropcatch domains is astounding. We’ve known that bad guys buy expired domains to repurpose them, but the way in which they were used, and the amount of money actors are willing to spend wasn’t well understood.” said Renée Burton, VP of Infoblox Threat Intel.

"Expired domains can be a shortcut to both trust and traffic, making dropcatch domains a higher risk than the average newly-registered domain.”

Huge investment

One investigation uncovered a threat actor dubbed Sable Squirrel, estimated to have invested more than $7 million in acquiring over 10,000 expired domains now used for illegal streaming, online gambling, and malware distribution.

Most support a large Asian sports piracy operation whose sites look like consumer streaming products. These offer live football, match schedules, chat rooms, mirrors, and mobile promotion.

However, Infoblox said it's not the streaming that's the real business: it's the betting platforms that the group appears to control.

"A subset of these same domains also operate as malware C2. We identified over 31,000 malware samples connecting to Sable Squirrel domains, including Quasar RAT, AsyncRAT, DCRat, NanoCore, Remcos, njRAT, and samples carrying HiddenTear ransomware signatures," said the researchers.

"In such cases, a human visitor sees a live football streaming site while an infected device uses the same domain as a control channel."

Meanwhile, other threat actors are taking over well-known malicious domains that are inserted into compromised websites. Researchers uncovered one actor delivering potential victims to SocGholish, the notorious 'fake update' infrastructure which was the target of Operation Endgame in June 2026.

Tracked as Shady Squirrel by Infoblox Threat Intel, the group delivered malware through scareware and call centers before partnering up with SocGholish’s operator TA569 in July.

"We see around 65,000 new dropcatch domains every day. Some of them are legitimate registrations picking up domains people are going to use for commercial or personal use. But an awful lot are grabbed for grey to black purposes," the researchers said.

"Any way you slice it, the risk posed by dropcatch domains is significant, arguably greater than that of newly registered domains."

FOLLOW US ON SOCIAL MEDIA

IBM and OpenAI have partnered up to help customers deploy AI at scale for core business operations.

Under the deal, IBM will create a dedicated OpenAI Practice, training thousands of consultants and engineers for expert-level certifications through the OpenAI Partner Network.

OpenAI models and products will be combined with IBM Consulting technology and expertise, the duo said, with the first areas to be targeted including financial services, government, telecommunications, and financial services.

“The organizations pulling ahead with AI are the ones turning it into a trusted part of how their business operates,” said Denise Dresser, former chief revenue officer at OpenAI.

“IBM Consulting and OpenAI are helping organizations make that shift, combining deep transformation expertise to deploy AI that is secure, operational, and aligned with real business priorities.”

Driving adoption rates

According to the two firms, the partnership will focus on applying AI to specific business industry processes, while also modernizing applications, redesigning back-office workflows, and strengthening cybersecurity.

"We’re starting with the business use case,” said Michael Healy, managing partner of offerings, assets and generative AI at IBM Consulting. “Then we bring IBM’s process transformation and industry expertise together with OpenAI’s models to change how that work actually gets done.”

OpenAI frontier models like GPT-5.6 and products such as Codex and ChatGPT Work will be embedded into IBM’s AI platform for delivering consulting services, IBM Consulting Advantage.

This will allow operating procedures and workflows to be analyzed to identify inefficiencies and help teams automate and streamline work using AI across finance, procurement, customer operations, and HR.

Tech modernization

As part of the deal, IBM said the duo also plan to help clients modernize legacy applications and accelerate software development by bringing together OpenAI Codex and ChatGPT Work with IBM’s industry, technology and domain expertise, including integration.

Here, the aim is to help organizations simplify engineering processes and speed up the delivery of new digital products and services.

“Codex as a coding assistant will accelerate the application development modernization cycle in combination with IBM Consulting Advantage’s coding harness and OpenAI’s models,” said Healy.

Following IBM’s participation in the OpenAI Daybreak Cyber Partner Program, the duo will expand their collaboration on cybersecurity.

This will involve combining OpenAI frontier AI capabilities with IBM Autonomous Security. This is an agentic AI service designed to deliver coordinated decision-making, response, and intelligence for cyber professionals.

“While enterprises are rapidly investing in AI, they are looking for practical ways to apply it across their core operations to deliver measurable business outcomes and create new commercial models,” said Andy Baldwin, global senior vice president, IBM Consulting.

“The challenge is not access to AI technologies — it’s integrating AI securely and at scale into complex enterprise environments and workflows. By embedding OpenAI’s technology with IBM Consulting’s AI assets, industry solutions, and cybersecurity capabilities, we can help clients accelerate secure, AI deployments at scale.”

FOLLOW US ON SOCIAL MEDIA

UK small businesses have an increasingly optimistic outlook on the potential of AI tools, new research suggests, but bold leadership will be required to ensure successful deployment.

A study published by Dell Technologies shows two-thirds (66%) of UK SMBs now view AI as a “route to growth”. More than half (56%), meanwhile, told the firm that AI could give their business a competitive advantage moving forward.

This optimism is driving investment, the research found, with funding primarily aimed at unlocking productivity gains, improving customer experience, and improving decision making.

Brian Horsburgh, UK small business country manager at Dell Technologies, told ITPro the study shows AI is having a marked positive impact for SMBs, particularly in terms of allowing workers to sharpen their focus on specific tasks.

“One of the realities of working in a small business is that everyone wears multiple hats. Many people find themselves spending hours on admin, reporting, marketing or customer queries, even though those aren't the things that inspired them to start or join the business,” he told ITPro.

“AI can help take some of that work off their plate, giving them more time to focus on customers, creativity, and growth.”

Pulling ahead of the pack

According to Dell Technologies, a small group of “AI front runners” has emerged among UK small businesses.

Representing around 7% of the firms surveyed, these report the strongest gains from using AI, allowing them to free up six or more hours each week with the technology.

Horsburgh told ITPro that a recurring theme with these front runners is having a clear cut strategy and not rushing headlong into adoption for the sake of it.

Indeed, these companies are more likely to have dedicated leaders championing AI than their slower-moving competitors.

More than half (52%) have “clearly defined specific AI use cases”, the study found, often starting with small examples before moving to more complex use cases as they mature.

“Having the right strategy is critical. Our research shows that those who are gaining the most from AI today say that AI use is coming from leadership as a priority,” Horsburgh said.

“Adopting AI means identifying ways it can help you do more of what helps your business grow and in fact, the number one piece of advice that came through from businesses of all sizes is to start small and iterate.”

Horsburgh said the findings highlight the need for dynamic, open-minded leadership when it comes to AI adoption. Initial use cases are key, he noted, but employee training and cultivating an “AI-friendly culture” are equally important.

“Those three pieces are critical to building a culture that drives benefits,” he told ITPro. “The outcomes that SMBs are seeing are positive for staff: more time to think strategically, improving decision making, improving customer experience, and having the right environment can help teams see those benefits sooner.”

The SMB agility dynamic

Dell Technologies’ study emphasized that small businesses have an inherent advantage over larger organizations when it comes to adoption.

First and foremost, they’re typically more agile and flexible with how they approach the technology – and the report noted that is now translating into higher rates of success.

This advantage isn’t lost on large businesses, either, Dell Technologies found. Nearly two-thirds (61%) of larger firms said greater flexibility is a key advantage for SMBs. Similarly, they have “more focused business needs” and are less risk averse.

SMBs also recognize these advantages, according to Dell Technologies, and are less likely than their larger counterparts to cite key barriers, such as lengthy approval processes, employee resistance to change, or fear of failure.

“Smaller businesses can have an advantage because they're often more flexible and can move faster than larger businesses, so starting with one or two key opportunities can potentially have an outsized impact.” Horsburgh told ITPro.

FOLLOW US ON SOCIAL MEDIA

Tricentis has announced the appointment of Erika Dean as chief information security officer (CISO), as the software quality specialist moves to strengthen its enterprise and product security capabilities.

Dean is tasked with leading Tricentis’ global security strategy, with responsibility for enterprise cybersecurity, compliance, and product security.

A seasoned industry veteran, she joins the company with more than two decades of cybersecurity leadership experience across the financial services and technology sectors. Previously, she built and led security programs at Robinhood and Capital One.

As CISO, Dean will work across the business to develop its enterprise and product security programs, with a focus on strengthening software supply chain security, as well as cyber resilience and governance across its portfolio.

Her remit also includes supporting the Tricentis Agentic Quality Engineering Platform as AI becomes increasingly embedded in the software development lifecycle.

“Cybersecurity is foundational to the trust our customers place in Tricentis,” said Tricentis CEO Kevin Thompson in an announcement. “Erika's expertise in building enterprise security programs and her customer-centric approach make her the ideal leader to drive Tricentis’ continued efforts to deliver innovative software quality solutions that enable customers to move faster without compromising security or governance.”

Dean’s appointment comes as businesses face increasingly sophisticated cyber threats and greater scrutiny around AI adoption, software supply chains, and the protection of sensitive data.

Tricentis eyes resilience gains

Austin-headquartered Tricentis develops software quality and testing technology designed to help enterprises automate and accelerate software development and release processes.

Its Agentic Quality Engineering Platform uses AI to automate software quality processes across enterprise application environments.

Tricentis said its latest leadership appointment reinforces its ongoing efforts to maintain a robust security program while helping customers deliver software securely.

Commenting on her new role, Dean said the growing use of AI in software development is changing the role security teams must play in protecting both businesses and their customers.

“As AI continues to transform the way software is built, security must extend beyond protecting our own organization to earning our customers’ trust in the software running in their critical environments,” she explained.

“Tricentis sits at a critical point in the software development lifecycle, where quality, security and governance all work together.

“I’m excited to build on the company’s strong security-first culture and partner with our customers to help them navigate the evolving cybersecurity landscape with confidence.”

FOLLOW US ON SOCIAL MEDIA

AMD is planning to offer up to £5 million in support for a new lab in Oxford carrying out research into frontier AI.

The company has signed a memorandum of understanding to provide compute resources for researchers at the British Open-ended Learning and Discovery Lab (BOLD) during the lab's first eighteen months in operation.

BOLD is led by the University of Oxford, and brings together researchers affiliated with the University of Oxford, University College London, and Imperial College London.

They, together with AMD researchers, plan to explore initiatives spanning compute access, software optimization, research collaboration and open-source development.

Work could include experiments to explore model training techniques beyond backpropagation, discovery using multi-agent systems and adaptive physical AI.

The lab will be built around three research aims - new learning algorithms, human-centered AI and embodied systems such as robots - and will translate breakthroughs into open source tools and encourage commercial spin-outs.

“Through our memorandum of understanding with the BOLD lab, AMD is exploring bringing together advanced computing, open software and technical expertise to help researchers pursue ambitious new frontiers in AI," said Thomas Zacharia, senior vice president, global public sector and strategic partnerships at AMD.

"By expanding access to compute and supporting open academic research, our aim would be to help BOLD researchers accelerate experimentation and contribute to the UK’s leadership in AI innovation and scientific discovery.”

AMD to provide underpinning infrastructure

AMD plans to provide support through infrastructure that will be maintained by the company itself, along with cloud service providers, public-sector partnerships and hardware contributions.

Phase 1 of the research program will focus on broad exploratory experiments, with later phases following up select research directions and validating the most promising ideas.

The company said it will work with BOLD researchers across the AI technology stack, exploring novel ways to optimize AI workloads using its open software ecosystem.

The ecosystem envisaged by the MOU would be built on permissively licensed open-source projects such as AMD ROCm software. AMD would help BOLD researchers publish research software, benchmarks, and evaluation tools.

"BOLD aims to advance a more open, efficient and human-centered path for AI through collaboration between academia and industry. Working with AMD, the University of Oxford would combine world-leading research with computing technology to test ideas, learn quickly and translate breakthroughs into applications," said associate professor Jakob Foerster, of the University of Oxford's Department of Engineering Science.

"This support would help BOLD develop approaches to learning, human-AI collaboration and embodied intelligence while expanding open access to computing resources that benefit science, healthcare, education, transportation and public services globally.”

UK pledges continue

The MoU marks the latest in a number of UK-related pledges made by AMD in recent months.

In June, for example, the chipmaker announced plans to invest £2 billion in the UK to drive AI innovation and research. This also saw AMD pledge to expand access to compute resources for enterprises across the country.

Along with University College London-led research lab Science of Fundamental AI Research (SOFAIR), BOLD is to receive a share of £60 million in government funding over the next six years.

This funding forms part of the £1.6 billion UKRI AI Strategy, which aims to strengthen the UK’s leadership in AI over the next four years.

FOLLOW US ON SOCIAL MEDIA

Want people to actually use AI? Then make it more useful according to Airbnb CEO Brian Chesky.

Speaking during a recent Power Players podcast appearance, Chesky said people are still using AI as an "interactive search product" but predicted further progress when it comes to interfaces, rather than just typing.

"I think part of it's a narrative issue that we're not talking about AI correctly," he said. "But part of it is we need to actually be developing more products that just regular people can use and say, 'I love AI because AI allows me to have a doctor on demand and I can't have that. I can't afford that.' And so I think we need more regular things."

Given Airbnb’s market, Chesky noted that the focus on largely enterprise-related AI innovation could be holding back broader adoption. He noted that 159 out of 175 startups at Y Combinator, where Chesky is a board member, focused on enterprise AI rather than consumer products.

"Maybe a simple way of saying it is we're getting really, really deep on enterprise, but there's been very little progress on consumer AI," he added.

AI backlash

Chesky said that it's in Silicon Valley's interest to make consumer apps that improve people's lives in order to shift increasingly negative opinions about the tech, warning that "regular people in the United States do not like AI."

This year has seen the start of a mainstream backlash against the technology, perhaps inevitable given the hype, but concerns have also been boosted by rising prices and environmental concerns around data centres.

Survey after survey also suggests many people simply don't like AI. Research via YouGov showed two-thirds of people weren't confident in the way businesses use generative AI, in particular with customer service.

A poll by Pew suggested just 16% of Americans think AI will prove to be a positive tool for society.

Chesky's argument is that building actually useful products powered by AI could help to undercut some of that negative sentiment – and he believes that will happen.

The Airbnb chief predicted that within in the next two or three years there will be a "renaissance around consumer AI that is going to begin to change daily life."

AI booster

Chesky has long been positive about AI. In 2023, he compared the technology's impact to electricity and said it will "change everyone's life."

In the podcast, he noted there haven’t yet been any major changes to our daily lives on par with the internet or iPhone.

Chesky also called for moving beyond the chatbot model, saying later that year that it should "augment humanity in a positive way".

Airbnb bought GamePlannerAI in November 2023, and earlier this year said AI was the "best thing that ever happened to Airbnb," in part thanks to a third of customer service tickets now handled by the technology.

After its last round of results, that number had jumped to 45% – despite those previously mentioned concerns about AI driven customer service.

FOLLOW US ON SOCIAL MEDIA

As AI continues accelerating code production, developers should take a leaf out of the manufacturing industry’s book to automate testing processes safely.

That’s according to Ingo Philipp, VP of product engineering at UiPath. Speaking to ITPro, Philipp suggested that “lights-out” manufacturing-style processes could alleviate strain placed on teams facing a torrent of code.

Also known as ‘dark factories’, lights-out manufacturing techniques involve fully automated production lines, and are a tried and tested method in the manufacturing industry.

Japanese engineering firm FANUC has been running these autonomous factories for more than two decades, requiring minimal human input except for quality control and overseeing production.

For developer teams, Philipp noted the same techniques could be applied to software development. UiPath coined the term ‘dark testing factories’ in a recent whitepaper, noting that this could both speed up and secure software testing.

“We see that the demand for software testing is just exploding these days. That’s what we see day in, day out from our customers,” he told ITPro.

“At the same time, we also see that they are struggling with the demand. That means they can’t really keep up with this increased volume, increased velocity of software development,” Philipp added.

“That means there is a gap essentially between the demand and the supply. So that's the challenge. I think it has always been a challenge.”

Matching velocity

As Philipp noted, the concept comes amid the growing use of AI in software development. Figures from Stack Overflow’s most recent Developer Survey found 84% of devs have adopted AI, with more than half (51%) using these tools everyday.

With increased production velocity comes added risks, however. In its 2026 AI Engineering report, engineering analytics platform Faros, showed teams now report 54% more bugs per developer compared to pre-AI processes.

Similar research from GitLab found that the volume of code written using AI is surging, yet 92% of developers reported significant governance challenges.

With these issues expected to continue, creating an isolated, automated production pipeline in which code can be tested safely is critical, according to Philipp.

“Think of Codex, think of Claude, think of all the Copilots and Coworks out there,” he said. “They’re enabling our friends in software development to crank out more code faster than ever these days.

“We see our customers, our partners really struggling to keep up with this increased volume and velocity in software development with the traditional approaches they are using in their software testing.”

Automating the pipeline

UiPath’s contention is that with ‘dark testing factories’, AI agents can automate aspects of software testing before it has the potential to cause downstream issues.

In these settings, AI agents would weed out potential bugs while human developers focus more on quality control, security, and governance. Philipp added that testers will eventually “teach the factory to do the work”.

Automation in software testing is by no means a new trend, he noted. In this instance though, it’s a case of speeding things up in line with acceleration across the broader development lifecycle.

“It’s autonomous testing on steroids,” he told ITPro. “The goal here is to … do away with all the operational, repetitive, mundane testing work from the humans and elevate them into more ‘quality leaders’.”

This does raise the question of whether enterprises are simply adding more automation – and complexity – to an already complicated and highly-automated process.

Philip said, however, that automation in this context will still require robust human oversight. Customers adopting this approach are able to create strict guardrails on what tasks agents can perform, much like other areas in which agents are deployed and a human-in-the-loop approach is still vital.

One customer, he said, set boundaries around what actions agents can take without human intervention, reverting to testers in certain circumstances.

“Who is actually the quality owner? Who is the person that needs to provide clarity on that? This is not something new, because they already have it as part of their existing testing process. We just factored it in.”

Philip told ITPro that while customers are shifting toward this approach, it’s at a moderate pace. Long-term, he envisages a far more AI-fuelled testing setup but only a handful of companies have the capabilities right now to achieve this.

“I want to be brutally honest here, almost none of our customers are there at the moment,” he said. “None of our customers [have] achieved the holy grail of making testing fully autonomous. I don’t think it is achievable in the short-term and mid-term.”

FOLLOW US ON SOCIAL MEDIA