Meta says it fixed an AI support flaw used to take over Instagram accounts
The facts
- Meta said it fixed a security flaw in its AI-powered support assistant after attackers used it to hijack Instagram accounts.
- The reported method involved getting Meta's support chatbot to link an attacker-controlled email address to a target Instagram account and then using password reset tools to take control of the account.
- Multiple high-profile Instagram accounts were reported as affected, including the Obama-era White House account, Sephora, and the U.S. Space Force chief master sergeant's account.
- Users on Reddit and X reported a wave of account hijackings over the weekend, indicating the issue was not limited to a few prominent accounts.
- Several reports said attackers did not need prior access to a victim's email account to carry out the takeover.
- Videos and screenshots showing the technique circulated on social platforms and messaging channels, helping expose how the chatbot could be used in account takeovers.
- What remains unresolved is the full scale of the incident: reports say multiple accounts were affected, but the total number has not been clearly established in Meta's public statements cited by these outlets.
Context
How did the account takeover reportedly work?
Multiple reports say attackers prompted Meta's AI support chatbot to add a new email address to a target Instagram account. After that change, they could request a password reset and use it to gain control of the account Indian Express,India Today,Times of India,TechCrunch.
Which accounts were publicly identified as affected?
Outlets reporting on the incident identified the Obama-era White House Instagram account, Sephora, and the account of the U.S. Space Force chief master sergeant among the compromised profiles Guardian,Indian Express,India Today,Verge.
What has Meta said about the issue?
Meta said the issue had been resolved, and one company statement cited by outlets said it was securing impacted accounts. The coverage reviewed here does not establish a confirmed total number of affected accounts Guardian,Anadolu Ajansı,TechCrunch.
Common ground, and where they split
- Common ground
- A support tool meant to help users instead became a path to account takeover, exposing a real security failure that left people and institutions vulnerable while Meta has yet to clarify how widely the flaw was exploited.
- Where Left and Right split
- Less a disagreement than a question of emphasis: the harm borne by users caught in a wider wave of hijackings, versus the governance failure of letting automated support change recovery details without tighter security bounds.
How left and right read it
What stands out is who absorbs the fallout when a platform’s own support system becomes the route to takeover: not just prominent accounts, but a broader set of users reporting hijackings, with the full scale still unclear. When attackers can seize accounts without prior access to a victim’s email, the burden falls on people and institutions that rely on these accounts while collective protection is shown to be fragile.
The core issue is institutional restraint: a support assistant appears to have been too easily induced to alter account-recovery information and open the door to password-reset takeovers. That points to a failure to keep automated support bounded by clear rules, leaving users more exposed and raising basic questions about whether convenience was allowed to outrun disciplined account security.
The receipts — 72 sources
Wire services (1)
Other coverage (71)
A news briefing built to show both sides.
The day’s biggest stories in one short brief, built to show both sides and the competing values behind the headlines. Free in your inbox.