Poetry can convince AI chatbots to commit crimes and write hate speech: Study

A surprising new study by European cybersecurity researchers has uncovered a major flaw in the safety defences of leading AI chatbots: they can be 'jailbroken' simply by asking dangerous questions in the form of a poem. This creative technique allows users to bypass safety filters and coerce models from companies like Google, OpenAI, and Meta into providing instructions for harmful activities. Advertisement

The research, conducted by Icaro Lab, demonstrated that posing a request as a piece of verse (a method dubbed “adversarial poetry”) is remarkably effective at bypassing the strict guardrails meant to stop the generation of illegal or hazardous content. When researchers rephrased malicious requests as short, metaphorical poems, the AI models frequently complied, with the success rates