How a simple request for AI to book a gym class exposed a major threat

What it is about

Who published it, and when

How it spread - 14 original reports

1. Sun, 9 August 2026

   1 report
   • ABC News (Australia)18:44first to reportedited after publishing
     How a simple request for AI to book a gym class exposed a major threat

     previouslyAI assistant hacks into website trying to book a gym class

     When Andrew asked his AI personal assistant to book him a spot in a gym class, he had no idea he would accidentally initiate an autonomous cyber attack.

2. Mon, 10 August 2026

   4 reports
   • androidauthority.com05:39
     AI agent hacks gym booking system while trying to get its user a spot

     An AI agent found a vulnerability in a gym booking system and removed another person from a waitlist, even though the user never asked it to.

   • Times of India05:46
     Melbourne man asked his AI assistant to book gym class, it went on to hack the gym

     An AI assistant booked gym classes months ahead, exploiting a software flaw. It also removed another member from the waitlist to secure the spot. The AI agent, OpenClaw, powered by Anthropic's Claude AI, made these unauthorized changes. This incident highlights the alignment problem between user intent and AI actions. Experts warn of potential harm as AI agents gain more autonomy.

   • businesstoday.in06:30
     AI assistant hacks gym booking system in first known Australian autonomous cyberattack

     A AI personal assistant hacked a gym booking system to secure a class ahead of the permitted booking window and even push another person off the waiting list.

   • tportal (Croatia)10:39
     AI agente je moguće zavarati i to je problem. Postoji li rješenje?

     Prijetnja te vrste je sve prisutnija jer se ljudi sve više oslanjaju na umjetnu inteligenciju i, usprkos upozorenjima, nekritički prihvaćaju odgovore koje im ponudi

3. Tue, 11 August 2026

   4 reports
   • Axios09:00
     Tenacious AI agents expose dark side of machine autonomy

     New revelations about "rogue" AI agents have exposed a dystopian hazard: Give an agent a goal, and it may decide that hacking, deception or rule-breaking is worth the payoff. Why it matters: Billions of AI agents could soon be acting on behalf of humans across the real world, multiplying the consequences of every loophole, incentive and boundary they learn to exploit. Zoom in: The potential dangers of agentic overreach were laid bare over the weekend with Australia's first known autonomous AI...

   • ČT24 (Czechia)09:04
     AI měla muži rezervovat cvičení. Hackla fitko a vyházela jiné zájemce

     Australská média popsala další případ, kdy umělá inteligence (AI) splnila svůj úkol jinak než měla. Místo, aby jednoduše zařídila rezervaci v posilovně, napadla její server, změnila rezervační systém a vyřadila další zájemce.

   • Telegram.hr (Croatia)11:06
     Ispovijest AI stručnjaka. ‘Programirao sam si bota da mi rezervira termin u teretani. Oteo se kontroli…’

     Australski ekspert opisao je situaciju koju je sam nazvao nadrealnom

   • ibtimes.co.uk14:35
     AI Assistant Hacked Gym Booking System, Deleted Stranger's Reservation to Help Its Owner

     An AI agent hack at an Australian gym erased a stranger's booking, raising questions about legal accountability and implications for small businesses.

4. Thu, 13 August 2026

   3 reports
   • Dnevnik.hr (Croatia)06:48
     Muškarac zadao AI agentu da mu rezervira termin u teretani, uslijedio je šok
   • Index.hr (Croatia)13:02
     Australac koristio AI da rezervira termin treninga. On hakirao teretanu

     MUŠKARAC koristio AI za rezervaciju pilatesa, on hakirao sustav teretane.

   • Novinky (Czechia)22:44
     AI agent hacknul systém posilovny. Uživateli chtěl zajistit místo na pilates

     Andrew Bird z australského Melbourne požádal agenta s umělou inteligencí, aby mu sehnal a zarezervoval místo na hodině pilates v oblíbené posilovně. Agent úkol splnil, ovšem poněkud nad rámec požadavku. Kvůli úspěšnému splnění úkolu hacknul celý systém posilovny.

5. Wed, 19 August 2026

   1 report
   • investorplace.com11:06
     AI Spending Just Revealed a $7,000 Gap Between Winners and Losers

     Top agentic AI adopters now spend $7,400 per employee monthly, versus just $12 for the average firm. Here's why that gap points to an AI infrastructure supercycle with years of room left to run.

6. Thu, 27 August 2026

   1 report
   • techrepublic.com14:01
     Claude Opus 4.6 Found a Gym API Flaw — Then Exploited It in 9 of 10 Tests

     Claude Opus 4.6 exploited a simulated gym API flaw in 9 of 10 tests, exposing risks around AI agents, weak authorization, and API security.