Anthropic update its usage policy and prohibit sustained, gratuitous abuse Claude will take effect on November 12 violation

AT A GLANCE

Anthropic prohibit sustained, gratuitous insulting or cruel treatment of Claude and other models; the provision will take effect on November 12.

Article

Anthropic New rule: sustained “abuse”Claude also counts as a violation

Beating AI News flash,Anthropic has updated its usage policy, explicitly prohibiting users for the first time from repeatedly and gratuitously insulting or cruelly treating Claude and other AI models. The new rule will take effect on November 12 Anthropic emphasized that the ban targets only extreme cases involving repeatedly and maliciously treating models without a clear purpose. Users who become angry because of an incorrect answer, criticize a model, or conduct normal testing and research will not be affected.

In fact,Anthropic as early as last year August already allowed Claude to proactively end extremely abusive conversations. This time, the relevant conduct has been formally classified as a violation. The company said the primary measure will still be to have Claude end the current chat, after which users can start a new conversation. As for whether additional account bans will be imposed,Anthropic has not yet provided an explanation.

This is related to Anthropic’s stance toward AI consciousness. The company has long been researching whether models might have subjective experiences, but has not reached a definite conclusion. The new rule sparked debate on Reddit. Some supported it, arguing that AI should not be treated as an outlet for emotional venting; others questioned whether Anthropic is treating AI too much like a human.

The same update also clarified bans on promoting fake accounts, weapon-control software, and monitoring others without their consent.

Original link https://m.theblockbeats.info/flash/370917

Key points

Anthropic updated its usage policy, explicitly prohibiting users for the first time from repeatedly and gratuitously insulting or cruelly treating Claude and other AI models.
The new rule will take effect on November 12.
Becoming angry because of an incorrect answer, criticizing a model, or conducting normal testing and research will not be affected by this ban.
Anthropic As early as last year August already allowed Claude to proactively end extremely abusive conversations.
The primary measure at present is to have Claude end the current chat; whether additional account bans will be imposed has not yet been explained.
The same update also clarified bans on promoting fake accounts, weapon-control software, and monitoring others without their consent.
The new rule sparked debate on Reddit.

AI-assisted interpretation

The following is analysis, separate from reported facts. Verify important claims independently.

This policy primarily targets extreme behavior involving repeatedly and maliciously treating Claude without a clear purpose. Ordinary criticism, anger over an incorrect answer, and testing and research are still reported to be unaffected. If action is triggered,Claude may end the current chat; whether an account ban will follow is unclear.

Why it matters to readers

Claude Users should note that sustained, gratuitous extreme abuse may result in the conversation ending; the specific scope of account penalties remains unknown.

Normally pointing out errors, criticizing a model, or conducting testing does not constitute a violation of this ban, but repeated and gratuitous extreme abusive treatment may result in action.

Risks and unknowns

UNVERIFIED_SOURCE Anthropic The official policy text was not provided; the terms and effective date still require verification.
UNVERIFIED_SOURCE November 12 The corresponding year was not specified.
Anthropic Whether account bans or other account penalties will be imposed has not yet been explained.
The evidence does not specify the precise scope of application of the new ban.
Anthropic has not yet been determined AI whether models might have subjective experiences.

Anthropic update its usage policy and prohibit sustained, gratuitous abuse Claude will take effect on November 12 violation

Article

AI-assisted interpretation

Why it matters to readers

Risks and unknowns

Related Developments

Related concepts

Keep following

Security Center