Microsoft touts new Copilot features in Excel, but says you shouldn’t use them if you want accurate results
The tech giant warned against using the tool for “tasks with legal, regulatory, or compliance implications”
Sign up today and you will receive a free copy of our Future Focus 2025 report - the leading guidance on AI, cybersecurity and other IT challenges as per 700+ senior executives
You are now subscribed
Your newsletter sign-up was successful
Microsoft quietly rolled out new AI features for Excel that allow users to generate formulas using the tech giant’s Copilot AI tool, but warned against using them if you're concerned about accuracy.
As detailed in a post on the Microsoft 365 Insider blog, the new ‘COPILOT’ function for the spreadsheet software allows users to automate tasks and skip the drudge work of filling out a myriad of cells and tinkering with formulas.
“It can be painful and time-consuming to wrangle data, summarize feedback, categorize information, and brainstorm ideas,” the blog post reads. “The new COPILOT function in Excel for Windows and Excel for Mac is here to save time and supercharge your workflows.”
Using natural language prompts, users can simplify this experience, Microsoft said. Example formulas touted by the firm specifically highlighted text summarization, using ‘=COPILOT("Summarize this feedback", A2:A20)’
Others included AI-generated text for product descriptions, using ‘=COPILOT("Create a description for this product based on its specs", B2:B8)’
The feature also works alongside existing Excel functions, and can be used inside formulas such as LAMBDA, SWITCH, WRAPROWS, and IF.
“You can use results from other formulas as part of its prompt,” the blog post reads. “This makes it easy to add AI features to your spreadsheets without changing how they are set up.”
Sign up today and you will receive a free copy of our Future Focus 2025 report - the leading guidance on AI, cybersecurity and other IT challenges as per 700+ senior executives
Excel AI features come with a big catch
There is one big caveat, however. In a separate support page, Microsoft warned users not to use the new AI function “for any task requiring accuracy or reproducibility”.
The support page also advised users not to use the Copilot function for “tasks with legal, regulatory, or compliance implications”.
“Avoid using AI-generated outputs for financial reporting, legal documents, or other high-stakes scenarios,” it reads.
Simply put, users will need to keep close tabs on the feature to ensure that it’s not producing false or misleading outputs.
“Its output should be reviewed and validated for accuracy, especially for critical business decisions or reports,” the original blog post states.
Microsoft also highlighted data privacy considerations when using the new feature, telling readers that data sent through the Copilot function is “never used to train or improve” its AI models.
“The information you input remains confidential and is used solely to generate your requested output,” the company said.
Taking the misery out of spreadsheets
Of course, this is a pilot scheme for selected Windows Insiders and is nowhere near full roll-out. Microsoft isn’t alone in touting AI as a game changer for spreadsheets, however.
In June, Perplexity also launched a tool aimed at automating spreadsheet drudge work. Dubbed Perplexity Labs, the tool allows users to craft spreadsheets and reports using natural language prompts.
The Labs tool can write and execute code to structure datasets, for example, or apply formulas and create charts based on simple user prompts.
Startups are also jumping on the bandwagon in this domain. This includes Sourcetable, a US-based startup developing what it claims is the world's first “self-driving spreadsheet”.
The autonomous spreadsheet tool also draws on natural language prompts - either through voice commands or keyboard - to carry out tasks on behalf of the user.
Make sure to follow ITPro on Google News to keep tabs on all our latest news, analysis, and reviews.
MORE FROM ITPRO
- Microsoft says AI is finally having a 'meaningful impact' on developer productivity
- Excel for the web just got a big makeover – and it’s a lot easier to use
- Microsoft CEO Satya Nadella says Excel changed the game for enterprises in 1985 – he’s confident AI tools will do the same

Ross Kelly is ITPro's News & Analysis Editor, responsible for leading the brand's news output and in-depth reporting on the latest stories from across the business technology landscape. Ross was previously a Staff Writer, during which time he developed a keen interest in cyber security, business leadership, and emerging technologies.
He graduated from Edinburgh Napier University in 2016 with a BA (Hons) in Journalism, and joined ITPro in 2022 after four years working in technology conference research.
For news pitches, you can contact Ross at ross.kelly@futurenet.com, or on Twitter and LinkedIn.
-
Salesforce targets telco gains with new agentic AI toolsNews Telecoms operators can draw on an array of pre-built agents to automate and streamline tasks
-
Four national compute resources launched for cutting-edge science and researchNews The new national compute centers will receive a total of £76 million in funding
-
Claude Code flaws left AI tool wide open to hackers – here’s what developers need to knowNews The trio of Claude code flaws could have put developers at risk of attacks
-
AI isn’t killing DevOps, you’re just using it wrongNews New research indicates that enterprises with mature DevOps processes are gaining the most from AI adoption
-
‘AI is making us able to develop software at the speed of light’: Mistral CEO Arthur Mensch thinks 50% of SaaS solutions could be supplanted by AINews Mensch’s comments come amidst rising concerns about the impact of AI on traditional software
-
Claude Code creator Boris Cherny says software engineers are 'more important than ever’ as AI transforms the profession – but Anthropic CEO Dario Amodei still thinks full automation is comingNews There’s still plenty of room for software engineers in the age of AI, at least for now
-
Anthropic Labs chief Mike Krieger claims Claude is essentially writing itself – and it validates a bold prediction by CEO Dario AmodeiNews Internal teams at Anthropic are supercharging production and shoring up code security with Claude, claims executive
-
AI-generated code is fast becoming the biggest enterprise security risk as teams struggle with the ‘illusion of correctness’News Security teams are scrambling to catch AI-generated flaws that appear correct before disaster strikes
-
‘Not a shortcut to competence’: Anthropic researchers say AI tools are improving developer productivity – but the technology could ‘inhibit skills formation’News A research paper from Anthropic suggests we need to be careful deploying AI to avoid losing critical skills
-
A torrent of AI slop submissions forced an open source project to scrap its bug bounty program – maintainer claims they’re removing the “incentive for people to submit crap”News Curl isn’t the only open source project inundated with AI slop submissions