Anthropic banned insulting Claude

Anthropic has made changes to its policy, prohibiting insults directed at the neural network Claude and its use for propaganda and weapons development.

This was reported by The Verge, citing sources.

"Anthropic is making changes to its service usage policy for the first time in over a year to reflect new and extremely risky cases of abuse, including election interference, weapons development, surveillance, as well as use in healthcare and finance," the publication states.

The main response mechanism to rude treatment remains the neural network terminating the dialogue with the user who insulted it. However, it has not been specified whether a new sanction in the form of account blocking will be introduced.

The company added that the updated rules will only be applied in extreme cases where users repeatedly exhibit cruelty and disrespect toward the model.