Anthropic bans baseless insults against Claude

TVS News AI, editorially supervised, presents Anthropic bans baseless insults against Claude.

Anthropic has published new guidelines for its AI chatbot Claude. Among other things, they now explicitly prohibit users from baselessly insulting and hurting the model.

What initially sounds unusual – after all, we're talking about software here – marks a small tightening in the tone of the AI industry. heise online reports this, citing the updated rulebook.

For you, this means: If you use Claude regularly, it's worth taking a look at the revised guidelines. What was previously silently tolerated may have consequences in the future.

What exactly has Anthropic changed?

Specifically prohibited is baselessly insulting and hurting the chatbot – heise online highlights this from the new guidelines. Which other behavioral rules the update includes and what sanctions apply for violations is not clear from the available report.

Why an AI provider regulates the tone

That a company like Anthropic codifies the tone of interaction with its AI in rules is new. It fits a trend: respectful treatment of language models is increasingly becoming a prerequisite – comparable to how people treat each other. How strictly the new rules will actually be enforced remains open for now, as does whether other providers will follow suit.