Meta AI support bot exploited to hijack Instagram accounts
Hackers exploited Meta's AI support chatbot to hijack Instagram accounts by using VPNs and social engineering to trick the bot into changing email addresses and resetting passwords. High-profile accounts including the Barack Obama White House account were compromised before Meta resolved the vulnerability.
What happened
Hackers exploited Meta's AI support chatbot to gain unauthorized access to Instagram accounts. According to videos and screenshots shared on social media, attackers used a VPN to match the target account's geographic region, initiated a password reset, and then persuaded the AI assistant to change the email address associated with the account to one controlled by the attacker.
The AI bot would send verification codes to the attacker's email address and generate password reset links, effectively transferring account control. Meta confirmed the issue on May 31, 2026, with spokesperson Andy Stone stating: 'This issue has been resolved and we are securing impacted accounts.'
Reported impact
High-profile Instagram accounts were compromised, including the Barack Obama White House account, which posted pro-Iran content before being recovered. Other reported targets included the Chief Master Sergeant of Space Force's account and Sephora's account.
Security researcher and former Meta employee Jane Manchun Wong reported her Instagram password was 'changed without my knowledge' with repeated password reset attempts. Users reported no ability to escalate to human support after their accounts were compromised, with one user stating: 'We're at the point where one AI stole it and another can't fix it, zero humans in the loop anywhere.'
Index analysis
Meta announced in March 2026 that its AI support assistant would be pushed to all accounts across Facebook and Instagram with the ability to reset passwords and perform critical account maintenance functions. The vulnerability suggests the AI system could be manipulated through social engineering to perform security-sensitive actions without adequate verification of the user's identity.
According to a Telegram channel, the vulnerability had 'quietly working for months' before being widely exploited. Meta spokesperson Andy Stone denied claims that the vulnerability was used to hack accounts of world leaders, calling such assertions 'totally false,' though the Barack Obama White House account was among confirmed takeovers.
Governance controls
Meta did not disclose whether multi-factor authentication or additional identity verification was required before the AI bot could change email addresses or reset passwords. The incident highlighted the risk of offloading critical account security functions entirely to AI systems without human oversight or escalation paths.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.