Wiz CTO speaks out amid confusion over Snowflake-GitHub Copilot flaw
Did Wiz spot an AI-written bug? Maybe not, but it did catch a five-day-old flaw
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.
Sign up today and you will receive a free copy of our Future Focus 2026 report - the leading resource for IT decision-maker insight on priorities and investment areas in AI, security and more.
“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
Follow ITPro on Google News and add us as a preferred source to keep tabs on all our latest news, analysis, views, and reviews.
You can also follow ITPro on LinkedIn, X, Facebook, and BlueSky.
Freelance journalist Nicole Kobie first started writing for ITPro in 2007, with bylines in New Scientist, Wired, PC Pro and many more.
Nicole the author of a book about the history of technology, The Long History of the Future.
-
What is AI insurance?In-depth The introduction of any new technology brings new risks, and AI's rapid expansion into the workplace is no exception. In response, a distinct category of insurance coverage has emerged
-
The CISO now owns physical security. Here’s what that means for the channelIndustry Insights Physical security budgets have moved to CISOs, and partners must adapt to this important shift
-
'The goal for this year will be to automate all security processes': Google Cloud is betting on Wiz to usher in a new era of AI securityNews Wiz wants to deploy its agents for continuous penetration testing, and in Google it’s found a parent company that can achieve this vision at scale
-
Microsoft patches six zero-days targeting Windows, Word, and more – here’s what you need to knowNews Patch Tuesday update targets large number of vulnerabilities already being used by attackers
-
An AWS CodeBuild vulnerability could’ve caused supply chain chaos – luckily a fix was applied before disaster struckNews A single misconfiguration could have allowed attackers to inject malicious code to launch a platform-wide compromise
-
Shai-Hulud malware is back with a vengeance and has hit more than 19,000 GitHub repositories so far — here's what developers need to knowNews The malware has compromised more than 700 widely-used npm packages, and is spreading fast
-
GitHub is awash with leaked AI company secrets – API keys, tokens, and credentials were all found out in the openNews Wiz research suggests AI leaders need to clean up their act when it comes to secrets leaking
-
Organizations urged to act fast after GitHub Action supply chain attackNews More than 20,000 organizations may be at risk following a supply chain attack affecting tj-actions/changed-files GitHub Action.
-
Nearly a million devices were infected in a huge GitHub malvertising campaignNews Microsoft has alerted users to a malvertising campaign leveraging GitHub to infect nearly 1 million devices around the world.
-
'GitVenom' campaign uses dodgy GitHub repositories to spread malwareNews Security researchers have issued an alert over a campaign using GitHub repositories to distribute malware, with users lured in by fake projects.