{
  "id": 18149067,
  "title": "Anthropic to prohibit persistent abuse of Claude chatbot",
  "url": "https://urgent.news/2026/11/04/anthropic-to-prohibit-persistent-abuse-of-claude-chatbot",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-04T14:49:47.000Z",
  "source": {
    "name": "Arabian Post",
    "slug": "arabian-post",
    "url": "https://thearabianpost.com/anthropic-to-prohibit-persistent-abuse-of-claude-chatbot/"
  },
  "original_language": "en",
  "account": "Anthropic will prohibit sustained, pointless abuse of its Claude AI models starting November 12, expanding existing safeguards against hostile interactions that Claude can already terminate. The US company's updated policy targets extreme cases where Claude is repeatedly cruel without purpose, while allowing normal frustration over inaccurate answers, criticism, and legitimate testing. The restriction, announced on October 8, is part of broader policy revisions covering deceptive online campaigns, election interference, surveillance, weapons development, and applications that could impact people's safety or livelihoods. Anthropic described the prohibition as applying to persistent and unnecessary abusive or cruel behavior, emphasizing it does not penalize users who challenge responses, express dissatisfaction, explore disturbing fictional material, or conduct research on model behavior. The company has not set a numerical threshold for what constitutes a violation, but Claude already has a mechanism to end certain conversations with abusive users. Anthropic stated that terminating those exchanges would remain its primary means of enforcing the new rule, rather than introducing separate punishments for each objectionable message. The existing conversation-ending capability emerged from Anthropic's work examining the welfare of advanced AI systems, without concluding they have experiences or interests. The policy does not establish Claude's consciousness or legal protections, but sets a contractual boundary on how customers may interact with the system. The update does not affect model testing or research, which remains allowed. Changes also address harmful uses of Claude beyond individual conversations, including deceptive campaigns, election interference, and surveillance. Prohibited assistance now extends to software and components used to operate weapons, such as guidance systems and autonomous vehicle arming. Surveillance must not be used to decide or recommend criminal proceedings, and permitted activities include fraud monitoring, content moderation, journalism, and legal research. Human oversight and disclosure are required for consequential applications affecting health, finances, legal rights, or essential services. The policy applies to all Anthropic products and services, but Claude's conversation termination remains the main response to persistent abuse.",
  "summary": "Anthropic will prohibit sustained, purposeless abuse of its Claude artificial intelligence models from 12 November, formalising restrictions on hostile interactions that the chatbot can already terminate under existing safeguards. The US technology company said its revised usage policy targets extreme cases involving repeated cruelty towards its models without any discernible purpose. Ordinary…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}