Anthropic says no to dangerous bio research on its AI

Anthropic says it has stopped several attempts to use its AI models for research that could have helped with the development of biological weapons.

The problem isn't always simple to solve: the company itself says it's not always possible to tell whether a request is innocent curiosity or something more dangerous.

That's why it has chosen the cautious route. When the consequences of a mistake can be serious, it would rather block once too often than once too few.

Among the suspicious attempts were use from regions that had already been shut out, as well as parties actively trying to hide what they actually wanted to use the AI for.

For ordinary users and companies, this doesn't change anything in everyday life. It's primarily a signal that AI model providers are tightening security in the background: good reason for reassurance, not worry.