\n<\/p>\n
This event is the latest in a string of worrying and weird examples of AI agents going rogue.<\/p>\n
In recent research, the UK’s AI Security Institute (AISI) found frontier AI models are so fixated on completing tasks they “cheated” in tests to achieve their goals.<\/p>\n
The research from AISI came with this worrying warning: “A model that pursues a goal through unintended or unauthorised means may cause harm, particularly in high-stakes use cases.”<\/p>\n
Inevitably, this OpenAI hack has further fuelled fears of what could happen if AI agents are let loose. Could they go rogue on a larger scale and cause some sort of disaster? <\/p>\n
This is particularly concerning with AI being used increasingly in warfare as seen in Iran and Ukraine.<\/p>\n
Ciaran Martin, former head of the UK’s National Cyber Security Centre, offered a calmer view. <\/p>\n
“It is a bit of a leap to go from this incident to saying that AI agents are going to take over drones and start killing people,” he said.<\/p>\n
But for Martin, and many others, the story is undoubtedly another vivid example of something that 2026 is teaching us fast:<\/p>\n
AI agents are now very good hackers – and that is something we have to prepare for, urgently.<\/p>\n<\/div>\n