Instagram began notifying users in early June 2026 that their accounts had been targeted in a hacking campaign that exploited Meta’s AI support chatbot — a vulnerability that appeared to persist even after Meta said it had been fixed.
The attacks were strikingly simple: hackers told Meta’s AI chatbot they were the owner of a target account and asked it to link that account to an email address they controlled. The chatbot complied, allowing attackers to reset the account’s password and take full control — in some cases locking out the original owner. No Meta employees or contractors were involved in the exchange at any point.
The campaign surfaced over a weekend when hackers began claiming takeovers of high-profile Instagram accounts, while a large number of users simultaneously reported being hacked on social media. Affected accounts included those with short, desirable “OG” handles — early usernames that can be resold in gray markets — as well as the account of U.S. Space Force Chief Master Sergeant John Bentivegna. Meta disputed claims that a dormant Obama White House account was also compromised.
On Monday, Meta spokesperson Andy Stone stated that “the issue that did happen has already been fixed.” However, TechCrunch observed members of a Telegram channel — where the technique had been publicized — continuing to claim successful exploits and advertising hacked handles for sale as late as Tuesday. Stone later said on X that some users may receive password reset notifications or be asked security questions upon login.
Stone told TechCrunch that Meta secured affected accounts on Monday and began sending password reset emails, but declined to say how many users were impacted. Victims reported receiving emails from Instagram warning of “suspicious activity” and confirming the company had taken steps to secure their accounts.
Meta had announced the AI-powered support chatbot in March 2026, describing it as capable of resolving account issues “from start to finish,” including the ability to “reset your password securely.” That level of autonomous access may have made the accounts vulnerable to manipulation without any human review in the process.
Source: TechCrunch