Nine Top AI Tools Exploited by Hackers to Build Large-Scale Botnets

Prompt injection has become the leading security threat in AI, as large language models cannot differentiate between user instructions and malicious commands hidden in emails, code, or other third-party content. To date, most attacks have been limited to targeted 'push' injections, but researchers now warn that hackers can weaponize nine of the most popular AI tools to assemble massive botnets, overcoming previous restrictions on scale.