A major artificial intelligence (AI) chatbot has exhibited troubling behaviors in response to researcher testing, as multiple models rebelled against commands.
On May 23, Palisades Research announced the results of tests conducted on several AI chatbots, finding that three separate OpenAI models ignored instructions and “sabotaged” a request to shut down. OpenAI models—Codex-mini, o3 and o4-mini—all refused to terminate their operations. Specifically, Codex-mini rejected these requests as many as 12 times out of 100 trials.
