Anthropic developed a defense against universal AI jailbreaks for Claude called Constitutional Classifiers - here's how it works.

AI firm Anthropic has developed a new line of defense against a common kind of attack called a jailbreak. A jailbreak tricks large language models (LLMs) into doing something they have been trained not to, such as help somebody create a weapon.

Anthropic has developed a filter system designed to prevent responses to inadmissible AI requests. Now it is up to users to test the filters.