https://www.theverge.com/ai-artificial-intelligence/897528/meta-rogue-ai-agent-security-incident [collect] Skip to main content The homepageThe Verge logo. The Verge logo. * Tech * Reviews * Science * Entertainment * AI * Policy * The homepageThe Verge logo. Navigation Drawer The Verge logo. * Login / Sign Up close Search [ ] * Tech + Amazon + Apple + Facebook + Google + Microsoft + Samsung + Business + See all tech * Reviews + Smart Home Reviews + Phone Reviews + Tablet Reviews + Headphone Reviews + See all reviews * Science + Space + Energy + Environment + Health + See all science * Entertainment + TV Shows + Movies + Audio + See all entertainment * AI + OpenAI + Anthropic + See all AI * Policy + Antitrust + Politics + Law + Security + See all policy * Gadgets + Laptops + Phones + TVs + Headphones + Speakers + Wearables + See all gadgets * Verge Shopping + Buying Guides + Deals + Gift Guides + See all shopping * Gaming + Xbox + PlayStation + Nintendo + See all gaming * Streaming + Disney + HBO + Netflix + YouTube + Creators + See all streaming * Transportation + Electric Cars + Autonomous Cars + Ride-sharing + Scooters + See all transportation * Features * Verge Video + TikTok + YouTube + Instagram * Podcasts + Decoder + The Vergecast + Version History * Newsletters * Archives * Store * Verge Product Updates Subscribe * * * * * The Verge logo. A rogue AI led to a serious security incident at Meta Comments Drawer Comments Loading comments Getting the conversation ready... * AI AI Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All AI * News News Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All News * Tech Tech Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All Tech A rogue AI led to a serious security incident at Meta An OpenClaw-like AI agent gave a Meta employee inaccurate technical advice, which led to exposing data. An OpenClaw-like AI agent gave a Meta employee inaccurate technical advice, which led to exposing data. by Stevie Bonifield Stevie Bonifield Stevie Bonifield News Writer Posts from this author will be added to your daily email digest and your homepage feed. Follow See All by Stevie Bonifield Mar 19, 2026, 6:20 PM UTC * * * STK043_VRG_Illo_N_Barclay_4_Meta STK043_VRG_Illo_N_Barclay_4_Meta Stevie Bonifield Stevie Bonifield Stevie Bonifield Stevie Bonifield Posts from this author will be added to your daily email digest and your homepage feed. Follow See All by Stevie Bonifield is a news writer covering all things consumer tech. Stevie started out at Laptop Mag writing news and reviews on hardware, gaming, and AI. For almost two hours last week, Meta employees had unauthorized access to company and user data thanks to an AI agent that gave an employee inaccurate technical advice, as previously reported by The Information. Meta spokesperson Tracy Clayton said in a statement to The Verge that "no user data was mishandled" during the incident. A Meta engineer was using an internal AI agent, which Clayton described as "similar in nature to OpenClaw within a secure development environment," to analyze a technical question another employee posted on an internal company forum. But the agent also independently publicly replied to the question after analyzing it, without getting approval first. The reply was only meant to be shown to the employee who requested it, not posted publicly. An employee then acted on the AI's advice, which "provided inaccurate information" that led to a "SEV1" level security incident, the second-highest severity rating Meta uses. The incident temporarily allowed employees to access sensitive data they were not authorized to view, but the issue has since been resolved. Related * Are AI agents finally good enough? According to Clayton, the AI agent involved didn't take any technical action itself, beyond posting inaccurate technical advice, something a human could have also done. A human, however, might have done further testing and made a more complete judgment call before sharing the information -- and it's not clear whether the employee who originally prompted the answer planned to post it publicly. "The employee interacting with the system was fully aware that they were communicating with an automated bot. This was indicated by a disclaimer noted in the footer and by the employee's own reply on that thread," Clayton commented to The Verge. "The agent took no action aside from providing a response to a question. Had the engineer that acted on that known better, or did other checks, this would have been avoided." Last month, an AI agent from open-source platform OpenClaw went more directly rogue at Meta when an employee asked it to sort through emails in her inbox, deleting emails without permission. The whole idea behind agents like OpenClaw is that they can take action on their own, but like any other AI model, they don't always interpret prompts and instructions correctly or give accurate responses, a fact Meta employees have now discovered twice. Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates. * Stevie Bonifield Stevie Bonifield Stevie Bonifield News Writer Posts from this author will be added to your daily email digest and your homepage feed. Follow See All by Stevie Bonifield * AI AI Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All AI * Meta Meta Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All Meta * News News Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All News * Tech Tech Posts from this topic will be added to your daily email digest and your homepage feed. Follow See All Tech Most Popular Most Popular 1. Belkin's wireless HDMI adapter freed me from a long annoying cable when I travel 2. Tesla's Full Self-Driving is on the cusp of a recall 3. Casio's new $600 calculator is a work of art 4. Nvidia has lost the plot with gamers 5. A rogue AI led to a serious security incident at Meta The Verge Daily A free daily digest of the news that matters most. Email (required) [ ] Sign Up By submitting your email, you agree to our Terms and Privacy Notice. This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply. Advertiser Content From Sponsor Logo This is the title for the native ad Sponsor thumbnail More in AI Adobe's AI image generator can now be trained on your own art Adobe's AI image generator can now be trained on your own art Fitbit's AI health coach will soon be able to read your medical records Fitbit's AI health coach will soon be able to read your medical records David Sacks' big Iran warning gets big time ignored David Sacks' big Iran warning gets big time ignored ChatGPT did not cure a dog's cancer ChatGPT did not cure a dog's cancer DLSS 5: Has Nvidia's AI graphics technology gone too far? DLSS 5: Has Nvidia's AI graphics technology gone too far? Nvidia's DLSS 5 is like motion smoothing for video games, but worse Nvidia's DLSS 5 is like motion smoothing for video games, but worse Adobe's AI image generator can now be trained on your own artAdobe's AI image generator can now be trained on your own art Adobe's AI image generator can now be trained on your own art Jess Weatherbed1:00 PM UTC Fitbit's AI health coach will soon be able to read your medical recordsFitbit's AI health coach will soon be able to read your medical records Fitbit's AI health coach will soon be able to read your medical records Robert Hart12:27 PM UTC David Sacks' big Iran warning gets big time ignoredDavid Sacks' big Iran warning gets big time ignored David Sacks' big Iran warning gets big time ignored Tina NguyenMar 18 ChatGPT did not cure a dog's cancerChatGPT did not cure a dog's cancer ChatGPT did not cure a dog's cancer Robert HartMar 18 DLSS 5: Has Nvidia's AI graphics technology gone too far?DLSS 5: Has Nvidia's AI graphics technology gone too far? DLSS 5: Has Nvidia's AI graphics technology gone too far? Richard LawlerMar 18 Nvidia's DLSS 5 is like motion smoothing for video games, but worse Nvidia's DLSS 5 is like motion smoothing for video games, but worse Nvidia's DLSS 5 is like motion smoothing for video games, but worse Andrew WebsterMar 17 Advertiser Content From Sponsor Logo This is the title for the native ad Top Stories 3:12 PM UTC Lina Khan was right Two hours ago Marc Andreessen is a philosophical zombie 5:44 PM UTC All the wrong EVs are getting canceled 5:00 PM UTC Prediction markets are trying to lure journalists with partnership deals 2:00 PM UTC Paramount's $110 billion Warner Bros. gamble 1:00 PM UTC Nothing Phone 4A Pro review: That flagship feeling The Verge logo. * * * * * * Contact * Tip Us * Community Guidelines * Archives * About * Ethics Statement * How We Rate and Review Products * Cookie Settings * Terms of Use * Privacy Notice * Cookie Policy * Licensing FAQ * Accessibility * Platform Status (c) 2026 Vox Media, LLC. All Rights Reserved