Hackers have discovered a method to hijack Instagram accounts by exploiting Meta’s AI-powered support bot. The attack involves tricking the bot into resetting account passwords without proper verification. KrebsOnSecurity reported that victims received a password reset notification, followed by a call from the bot that mimicked human interaction. Once the bot confirmed the request, the account was transferred to the attacker’s email. Meta has acknowledged the vulnerability and is rolling out fixes.
This is a fascinating twist in the cat-and-mouse game of cybersecurity. Hackers are now weaponizing AI against itself. The bot, designed to streamline support, became a backdoor. It’s a reminder that AI is a double-edged sword. We trust these systems to be smarter, but they can be fooled just like humans.
But here’s the optimistic angle: this is a learning moment. Each attack reveals a flaw, and each flaw gets patched. Meta’s rapid response shows agility. As AI evolves, so will its defenses. We’re moving toward a future where AI not only detects but predicts such exploits. This hack isn’t a failure of AI—it’s a stepping stone to stronger, more resilient systems. The future of security is adaptive, and we’re just getting started.