• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI
Report

Anthropic updates Claude usage policy: Bans 'sustained and needless abusive or cruel behavior' and relaxes political restrictions

A user says cybersecurity rules cover Claude Code and Computer Use; high-risk uses require human oversight only for consumer-facing outputs.

3 Sources, 1h ago, first seen 1h ago

TLDR

A writer says Anthropic's restriction on 'sustained and needless abusive or cruel behavior' was set to take effect November 12, 2026, and argues its 'model welfare' framing masks engineering and business concerns. Another user says the revised rules allow legitimate political discourse while barring deceptive or disruptive political uses, and add cybersecurity boundaries for Claude Code and Computer Use.

Combined views

—

3 Sources, first seen 1h ago

— likes— comments— saves— reposts

Combined views

—

3 Sources, first seen 1h ago

— likes— comments— saves— reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

3 Sources

Mσσɳ Qυҽҽɳ🌙👑@realkellyspearAnthropic's new "model welfare" policy isn't about protecting Claude from emotional harm. It is a strict engineering control designed to prevent RLHF data contamination and enterprise sycophancy. https://x.com/i/article/21085065731777658881h
Vaibhav Sisinty@VaibhavSisintyAnthropic updated Claude's usage policy land the political content change is the one that matters most Previously Claude had a blanket ban on all political and lobbying content. Users pushed back it was blocking legitimate policy research, civic education, and political writing Now only deceptive or disruptive political use is banned. Legitimate discourse is allowed Other changes: → New cybersecurity rules for agentic AI Claude Code and Computer Use are powerful enough now that clear boundaries on malicious use were needed → Law enforcement language simplified same rules, clearer writing → High-risk use cases like legal and financial AI now only need human oversight when outputs are consumer-facing, not internal The political content shift is a big deal. One of the most common complaints about Claude was that it refused to help with anything remotely political.1h
Sy An Ly@lysyan911@TokenGremlin Yep, until Anthropic ban your account because one of your silly prompt violate their models1h
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    3 Sources

    Mσσɳ Qυҽҽɳ🌙👑@realkellyspearAnthropic's new "model welfare" policy isn't about protecting Claude from emotional harm. It is a strict engineering control designed to prevent RLHF data contamination and enterprise sycophancy. https://x.com/i/article/21085065731777658881h
    Vaibhav Sisinty@VaibhavSisintyAnthropic updated Claude's usage policy land the political content change is the one that matters most Previously Claude had a blanket ban on all political and lobbying content. Users pushed back it was blocking legitimate policy research, civic education, and political writing Now only deceptive or disruptive political use is banned. Legitimate discourse is allowed Other changes: → New cybersecurity rules for agentic AI Claude Code and Computer Use are powerful enough now that clear boundaries on malicious use were needed → Law enforcement language simplified same rules, clearer writing → High-risk use cases like legal and financial AI now only need human oversight when outputs are consumer-facing, not internal The political content shift is a big deal. One of the most common complaints about Claude was that it refused to help with anything remotely political.1h
    Sy An Ly@lysyan911@TokenGremlin Yep, until Anthropic ban your account because one of your silly prompt violate their models1h
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet