Anthropic Sets New Rules on Abusive User Conduct Toward Claude

CNN Central News & Network-ITDC India Epress/ITDC News Bhopal: Anthropic Sets New Rules on Abusive User Conduct Toward Claude

Artificial intelligence (AI) company Anthropic has changed the rules governing the use of its AI chatbot Claude. Under the new use policy announced on October 8, 2026, users will be prohibited from repeatedly engaging in abusive or cruel behaviour toward the AI model without any reasonable cause. The company says the rule will apply only to serious cases in which users intentionally repeat such conduct.

Anthropic clarified that the restriction will not apply to every form of criticism. Expressing normal frustration, questioning responses, testing the model’s limits, conducting research, or writing creatively on serious and dark themes will not fall under the ban. According to the company, Claude may end a conversation in cases of sustained abusive interaction. For now, ending the conversation will remain the main course of action in such cases.

The new rules are not limited only to behaviour toward the AI. The updated policy also places clear restrictions on misleading political and commercial campaigns, deceiving people through fake accounts or counterfeit news websites, and influencing the electoral process. In addition, the rules have been tightened for uses such as developing software linked to weapons operation and monitoring people without their consent.

Conditions such as oversight by a qualified human expert have also been clarified for the use of AI advice in health, finance and other sensitive sectors. According to Anthropic, the new policy will take effect from November 12, 2026.

Hashtags: #ScienceTechnology #Bhopal #Claude #Anthropic #AI #AIChatbot #UsePolicy #AbusiveConduct #ArtificialIntelligence #AnthropicClaudeNewUse