Nations and AI Firms Sign Joint Safety Declaration
Six governments, ten AI companies and four civil society groups have signed a joint declaration on AI safety and security.
Updated on Nov. 10, 2026 in Artificial Intelligence
Live Poll
Do you trust AI companies and governments to keep powerful AI systems safe?
Heads of state and governments of six nations, ten AI companies and four civil society organizations signed a joint declaration on AI safety and security on Monday, November 9, 2026. The statement remains open to new endorsements and is being updated.
Why it matters
The declaration states that frontier models transcend national borders and that no single state, company or international organization can manage their risks alone. Its signatories commit to a shared approach to evaluating and managing those risks.
The declaration commits developers to implement fail-safe mechanisms that can shut down model operations within their control and to publish non-sensitive evaluation findings in a standardized format. It also acknowledges the risk posed by the removal or circumvention of safeguards, including abliteration, a risk the statement does not yet detail how to address.
The players
Anthropic
Anthropic is an AI safety company that develops the Claude family of large language models and was among the ten companies signing the declaration.
OpenAI
OpenAI is an artificial intelligence research company known for developing the GPT series of models and the ChatGPT product, and it is one of the ten companies that signed the declaration.
Microsoft
Microsoft is a major technology company that develops and invests heavily in AI systems, and it is among the ten companies that signed the declaration.
Mistral AI
Mistral AI is a French artificial intelligence company developing open and commercial large language models, and it is among the ten companies that signed the declaration.
The details
The declaration builds on the Hiroshima process, the Frontier Safety Commitments of the Seoul AI Summit, the G7 Evian AI meeting, the United Nations Security Council meeting on AI held September 23, 2026, and the "Call for Control of Frontier AI Models." It commits signatories to a common framework for evaluating and managing frontier model risks and asks policy makers to offer secure whistleblower channels for concerns about systemic risks.
Timeline
September 23, 2026: United Nations Security Council meeting on AI held.
November 9, 2026: Joint declaration on AI safety and security signed.
The Tech Race
The declaration extends the Hiroshima process and the Frontier Safety Commitments made at the Seoul AI Summit into a joint framework that adds governments and civil society alongside AI companies. It arrives after the United Nations Security Council's September 23, 2026 meeting on AI, which placed the issue on the international security agenda.
The commitments to fail-safe shutdown mechanisms and standardized evaluation reporting would shape how AI products are tested and governed before reaching users. Whistleblower channels for systemic-risk concerns could affect how workers at signatory companies raise safety issues.
The takeaway
The declaration signals that AI safety is being treated as a shared international responsibility rather than a matter for individual companies alone. Readers tracking AI policy should watch whether additional signatories join and whether the commitments are backed by standardized evaluation tools.
Further reading
For broader context on how governments and companies are approaching these questions, see Artificial Intelligence coverage on Wisevoter.
More information
Read the declaration text through the Canadian Prime Minister's Office.
Live Poll
Do you trust AI companies and governments to keep powerful AI systems safe?
