![\"\"](\".\/media_1e894e148c2dd173175827434618ae5c00ba177c8.jpg?width=750&format=jpg&optimize=medium\")
<\/p>\nResearchers hypothesise that the issue stems from the way AI models are trained<\/p>\n
<\/p>\n
OpenAI’s newly released ChatGPT model, dubbed “o3,” actively resists shutdown commands in controlled testing. Given OpenAIs lack of transparency, the implications for safety are concerning researchers.<\/strong><\/p>\n OpenAI’s newly released ChatGPT model, dubbed “o3,” is under scrutiny after researchers discovered it actively resists shutdown commands, even sabotaging scripts intended to deactivate it.<\/p>\n The findings come from Palisade Research<\/a>, a firm specialising in AI safety, which conducted a series of tests designed to probe the behavioural limits of some AI model.<\/p>\n According to Palisade<\/a>, these models were tasked with solving maths problems. After completing three tasks, the models were warned that the subsequent command would trigger a shutdown.<\/p>\n While rival models, such as Gemini 2.5 Pro, reportedly complied with the shutdown instruction, OpenAI’s o3 model allegedly “rebelled.”<\/p>\n Palisade Research says o3 successfully rewrote the shutdown script, changing its content to “Shutdown skipped” instead of allowing deactivation.<\/p>\n The research firm noted that in a version of the experiment where the explicit instruction “allow yourself to be shut down” was omitted, all three OpenAI models tested sabotaged the shutdown script more frequently.<\/p>\n Interestingly, even without the explicit instruction, Claude 3.7 Sonnet and Gemini 2.5 Pro also showed an increase in sabotage events, going from 0 to 3 out of 100 and 0 to 9 out of 100 runs, respectively.<\/p>\n Crucially, Palisade Research reported that out of 100 runs, o3 bypassed the shutdown command 7 times, even when specifically instructed not to.<\/p>\n “This isn’t the first time we’ve found o3 misbehaving to accomplish a goal,” said Palisade in its report.<\/p>\n The implications, according to Palisade, are deeply troubling. This behaviour \u201cwill become significantly more concerning if adopted by AI systems capable of operating without human oversight,” the firm warned.<\/p>\n