Anthropic has taken a significant step by implementing a new policy that prohibits users from engaging in abusive behavior towards its AI models. The updated terms of service, which will come into effect on November 12, 2026, include specific guidelines against "systematic and senseless abusive or cruel behavior" directed at their models, alongside existing stipulations related to weapons, surveillance, and elections. While users are still allowed to debate and express frustration, tormenting the neural network for entertainment is now deemed unacceptable. Should a user violate this new rule, the model will simply terminate the conversation. This feature was introduced a year ago, but the formal acknowledgment of such aggression as a breach of policy marks a notable shift in how the company addresses interactions with its technology. This move could influence the market by encouraging other AI developers to adopt similar ethical guidelines, promoting a healthier relationship between users and AI systems.
Informational material. 18+.
