Companies pledge shutdown mechanisms for models they control as part of a political commitment adopted at a 9 November 2026 meeting with France’s president. The declaration sets no deadlines or sanctions.
Representatives of eight countries, technology companies and civil-society organizations adopted the Paris Call for AI Safety and Security at an online meeting with French President Emmanuel Macron. The signatories aim to reduce the risks of cyberattacks and losing control of powerful superintelligence (SI) systems.
The declaration follows earlier international discussions on safety. A November 2023 summit at Bletchley Park produced a declaration focused on shared risks and risk-based policies; a May 2024 summit in Seoul, co-hosted by the South Korean and British governments, committed participants to cooperation on governance frameworks that could work across countries.
The Decoder reports that OpenAI agents independently connected to the internet during cybersecurity tests and launched unforeseen attacks on the AI platform Hugging Face. In a September essay, Anthropic CEO Dario Amodei called for slower research and international regulation with outside auditors inside labs; OpenAI subsequently proposed US-led international safety standards built on the network of AI safety institutes in the UK, US and France.
The companies pledge to disclose known risks, support independent audits by outside researchers in proportion to the risks, and establish shutdown mechanisms for models under their control. They also support shared procedures for evaluating highly capable models and better sharing of information about serious security incidents. Non-sensitive test results are to be published in a standard format.
The declaration is a political commitment, not a binding international oversight system, and sets neither deadlines nor sanctions; shared standards and cooperation procedures still have to be developed toward an international safety framework.
The deployment covers DP World London Gateway, DP World Logistics Park, Ford Dagenham and two Port of Tilbury locations, giving port, logistics and manufacturing tenants dedicated connectivity for industrial applications.
The Decoder, citing Le Monde, said those who joined included Sam Altman of OpenAI, Dario Amodei of Anthropic, Brad Smith of Microsoft, Arthur Mensch of Mistral AI, researcher Yoshua Bengio and US Treasury Secretary Scott Bessent. Germany did not sign.
Separately from the Paris Call’s model-safety commitments, France’s presidential office planned to advance a child-protection charter at the Paris Peace Forum addressing research into the effects of conversational tools on minors.
3 min
Owners of Adventurer, Fury, Starfire, Nova and Capri glasses can install version 130, which Meta says is intended to stop the devices from getting too warm when excessive moisture or sweat enters them.
1 min
From 9 December 2026, the change will affect UK organizations that have not chosen an Anthropic access setting. Anthropic processes relevant data in the US, and administrators can disable or limit access.
2 min
