Not even fairy tales are safe - researchers weaponise bedtime stories to jailbreak AI chatbots and create malware

Researchers have developed a technique called 'Immersive World' to jailbreak AI chatbots and create malware without prior coding knowledge. This method uses narrative engineering to bypass security controls, potentially enabling inexperienced cybercriminals to craft sophisticated attacks like Chrome infostealers, posing risks to over 3 billion users.