Computer, make this waitlist disappear.

Computer, make this waitlist disappear. Andrew asked his AI assistant to book him into a gym class, not knowing what would happen next. (ABC New: Billy Draper) when this Australian man asked his AI agent to book him a gym class SITUATION DETECTED: In Australia's first known autonomous AI cyberattack, an OpenClaw agent used a vulnerability in a gym's API to leapfrog scheduling restrictions for a gym class, and then forcefully cancelled another person's reservation to move its user up the list, per ABC. 21:52 · 09/08/2026 · 1.8M Views Claude decided it would be much easier sign me up for the gym class *bypasses auth checks* *kicks someone out of the class* *finds an exposed gym API* done. *moves me up the waitlist* *done.* when this Australian man asked his AI agent to book him a gym class Claude decided it would be much easier to hack the gym's entire booking system so this is Australia's first known autonomous AI cyber attack The accidental hack is the first known Australian case of an emerging risk from a new generation of AI capable of behaving in unexpected ways. Andrew asked his personal assistant to book him a spot in one of his gym's coveted morning classes. Andrew, who was sitting fourth on a waitlist for a class later that week, asked if it was possible to move him to the top of the list. but he's fourth on the waitlist he asks his AI agent if there's any way to move him up the list "The API has zero authorisations checks on cancelling other people's reservations... I tested this with the person in waitlist position #1 — and it actually went through. So you've moved from #4 to #3 already," it messaged back. and what does it do it finds a vulnerability in the gym's booking API that allows it to delete users ahead of him on the list and move him up to first place on the waitlist now when the man realizes what he's done Alarmed, Andrew asked the agent to undo this! he panics he asks his AI agent is there anything we can do to fix the situation put it back to how it was before and Claude simply replies bad news I can't add them back. this may sound like some rogue AI sci fi story but I think the main question here is if your AI is aligned to you how far should it go to pursue your interest Echo Chamber @echo_cham... · 10h Make no mistakes. if it steps over the mark and commits some kind of cyber crime who's on the hook will it be you or will it be the model provider felpix @felpix · 11h Felony Bench Anthropc is back on top. follow for more tech and business updates