AI red lines
In plain wordsLimits on AI that would be forbidden worldwide no matter what — for example AI that copies itself without permission or controls nuclear weapons — which campaigners want governments to agree on.
Proposed internationally agreed prohibitions on AI uses or behaviors judged too dangerous to permit under any circumstances, defined clearly enough to be verified and enforced. Proposals distinguish two families. Limits on uses would forbid, for example, AI that impersonates humans, AI-enabled mass surveillance, weapons that select and kill targets without meaningful human control, and handing AI the authority to launch nuclear weapons. Limits on behaviors would forbid systems that copy or upgrade themselves without authorization, engage in power-seekingpower-seeking, n.: The tendency of a goal-pursuing AI to gather resources, influence and options — and to resist being stopped — because these help with almost any goal it might have., mislead their own developers or give substantial help toward building weapons of mass destruction. The prohibited AI practicesprohibited AI practices, n.: Uses of AI that the EU bans outright as too harmful — such as manipulating people, social scoring, and most live facial recognition by police in public places. of the EU AI ActEU AI Act, prop. n.: The European Union's law on artificial intelligence, the first broad AI law anywhere: it bans some uses, sets strict checks for risky ones and puts duties on the makers of the biggest AI models. work as domestic red lines on uses, and companies' capability thresholdscapability threshold, n.: A level of dangerous ability — such as meaningfully helping someone make a biological weapon — that, once an AI model reaches it, obliges its developer to add stronger safeguards or hold the model back. are a voluntary analogue for behaviors.
The scientists urged governments to agree on AI red lines before systems could replicate themselves.
Origin
The International Dialogues on AI Safety, which brought together Western and Chinese scientists, proposed behavioral red lines in their Beijing statement of March 2024. The Global Call for AI Red Lines, led by the French Center for AI Safety (CeSIA) and The Future Society and launched at the UN General Assembly in September 2025, asked governments to agree on enforceable red lines by the end of 2026; by 2026 it listed more than 300 prominent signatories and 90 organizations.
Usage note
Supporters cite precedents such as the bans on biological weapons and blinding lasers. Skeptics doubt that behaviors such as deception can be defined or verified precisely, or that major powers will accept binding limits. As of October 2026 the campaign reported talks with diplomats from more than 20 countries but no formal endorsement by any government.
Related
- prohibited AI practicesprohibited AI practices, n.: Uses of AI that the EU bans outright as too harmful — such as manipulating people, social scoring, and most live facial recognition by police in public places.
- capability thresholdcapability threshold, n.: A level of dangerous ability — such as meaningfully helping someone make a biological weapon — that, once an AI model reaches it, obliges its developer to add stronger safeguards or hold the model back.
- loss of controlloss of control, n.: A situation in which people can no longer oversee, correct or switch off an AI system, and have no reliable way of getting that power back.
- Global Dialogue on AI GovernanceGlobal Dialogue on AI Governance, prop. n.: A yearly United Nations meeting, launched in 2025, where all countries — joined by companies, researchers and civil society — discuss how to cooperate on governing AI.
- Independent International Scientific Panel on AIIndependent International Scientific Panel on AI, prop. n.: A group of 40 independent experts set up by the United Nations in 2025 to give every country a shared, science-based picture of what AI can do, its risks and its benefits.
- AI pauseAI pause, n.: A proposed halt, temporary or indefinite, on building AI systems more powerful than today's, to give safety research and regulation time to catch up.
Sources
- Global Call for AI Red Lines (2025). Campaign statement and signatory list. red-lines.ai
Mentioned in
AI pauseAI pause, n.: A proposed halt, temporary or indefinite, on building AI systems more powerful than today's, to give safety research and regulation time to catch up., Independent International Scientific Panel on AIIndependent International Scientific Panel on AI, prop. n.: A group of 40 independent experts set up by the United Nations in 2025 to give every country a shared, science-based picture of what AI can do, its risks and its benefits., prohibited AI practicesprohibited AI practices, n.: Uses of AI that the EU bans outright as too harmful — such as manipulating people, social scoring, and most live facial recognition by police in public places., Global Dialogue on AI GovernanceGlobal Dialogue on AI Governance, prop. n.: A yearly United Nations meeting, launched in 2025, where all countries — joined by companies, researchers and civil society — discuss how to cooperate on governing AI., IAEA for AIIAEA for AI, phr.: The idea of an international agency for the most advanced AI, modeled on the UN's nuclear watchdog, which inspects countries' nuclear programs to keep them safe and peaceful., Statement on SuperintelligenceStatement on Superintelligence, prop. n.: A 2025 public statement, signed by scientists, celebrities and politicians, calling for a ban on building superintelligent AI until it is shown to be safe and the public supports it.