Anthropic bans ‘abusive or cruel behavior’ toward Claude
The Verge reports a usage-policy change that also adds rules on propaganda campaigns, surveillance and weapon development.
TL;DR
- The Verge reports that Anthropic’s updated usage policy bans abusive or cruel behavior toward Claude. [1]
- The same update adds rules aimed at propaganda campaigns, surveillance and weapon development. [1]
- The policy text is the company’s rule for users. It does not by itself show how often the rule will be enforced. [1]
The Verge reports that Anthropic has changed its usage policy to ban abusive or cruel behavior toward Claude. The headline is about conduct directed at the model, not only about what the model is allowed to say. [1] [1]
The same report says the update also adds rules addressing propaganda campaigns, surveillance and weapon development. Those are separate restrictions from the abuse clause, and The Verge presents them as part of the same policy revision. [1] [1]
A usage policy is a statement of what a company will treat as a violation. It is not a measurement of how many accounts are closed, and this brief does not have an enforcement tally. [1] [1]
The change was discussed at length on Hacker News. Readers can treat the policy categories as confirmed by The Verge’s account of the update, and treat any claim about Anthropic’s motives as commentary unless the policy text itself states it. [1,2] [1] [2]
Why it matters
Model providers are starting to write rules for how people treat the product, alongside the older rules for harmful uses.
Editor's note
The categories listed are from The Verge’s description of the policy update, not a line-by-line paste of Anthropic’s legal text.