Other appeals
People have been raising this for years.
The Accord did not start this conversation, and it does not claim ownership of it. Researchers, industry leaders and hundreds of thousands of members of the public have already spoken through the appeals below.
These are not Accord signatures, and they are not added together. Each initiative asks for something different, and one person may have signed several. There is no number here that represents unique people across all of them, so we do not print one. Every figure is what that organisation displayed on the date shown.
Center for AI Safety
Statement on AI Extinction Risk
700+ researchers and public figures · as shown on September 13, 2026
A single-sentence statement signed by many of the researchers and executives who built modern AI, including Geoffrey Hinton, Yoshua Bengio, Demis Hassabis, Sam Altman and Dario Amodei. Its purpose is to make it easier to say publicly that severe AI risks are worth taking seriously.
Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.
Curated rather than open: the form asks for title and affiliation, and listing is editorial. The unit of value is who signed, not how many. Their separate Take Action tool contacts federal and state representatives and is US-only.
Future of Life Institute
The Pro-Human AI Declaration
134,015 signatories · as shown on September 13, 2026
The largest public AI-safety declaration currently open for signature. It argues that AI should amplify human potential and protect dignity and liberty, rather than replace people and concentrate power in unaccountable institutions.
A broad coalition demands AI serve humanity, not the reverse.
Largest genuine public count in this field.
Future of Life Institute
Statement on Superintelligence
31,810 signatories · as shown on September 13, 2026
Calls for superintelligence development to be prohibited until two conditions are met: broad scientific consensus that it can be done safely and controllably, and strong public buy-in.
A prohibition on the development of superintelligence, not lifted before there is broad scientific consensus that it will be done safely and controllably, and strong public buy-in.
Future of Life Institute
Pause Giant AI Experiments
5,218 signatories · as shown on September 13, 2026
The 2023 open letter that put AI pause into mainstream debate. It also called for shared safety protocols, independent oversight, dedicated regulatory authorities, auditing and certification, and liability for AI-caused harm — ideas that substantially predate this Accord.
All AI labs should immediately pause for at least 6 months the training of AI systems more powerful than GPT-4.
Commonly cited as having tens of thousands of signatures. FLI's own open-letters index displayed 5,218 when checked on 13 September 2026. We show what the source shows.
PauseAI
PauseAI Statement and Campaign
2,858 signatories · as shown on September 13, 2026
An international movement building an organised political constituency for an AI pause, through local chapters, trained organisers, conferences and protests. Its strength is organised people rather than a signature count.
Governments should sign an international treaty implementing a pause on the training of the most powerful general AI systems, until we know how to build them safely and keep them under democratic control.
Their Theory of Change describes PauseAI as the activation layer that converts public concern into political pressure. The signature count is not the measure of the movement.
So why does the Accord exist?
Every appeal above asks you to agree with a specific position: that extinction risk is the right framing, that development should pause, that superintelligence should be prohibited, or that a particular set of values should govern AI.
Those are legitimate positions, and if one of them is yours, sign it, the links are right there.
The Accord asks something narrower and more basic: that governments take this seriously, build real oversight, and work together internationally. You can sign it while disagreeing about a pause. You can sign it while believing AI will be enormously beneficial.
Two other things are different here:
- The count can be checked rather than trusted. Every daily total is published, signed and linked to the previous day, and any signer can prove their own signature was counted. No other appeal in this field offers that. See how.
- The country is the unit. Every nation has its own page, its own count, and a source-linked record of what that government has actually done. Browse them.