Which AI labs commit to third party vetting by October 31?
Open13as of
Market pricing makes Microsoft the favorite at 23% across 13 tracked outcomes, as of .
What are the odds right now?
Each outcome shows its current market price — the market-implied probability it happens. Click an outcome for its full market page.
What moved the odds?
4 AI-matched news signals available across this event, newest first. Entry price is the called side when the news hit; the signed return marks it against the outcome’s latest price as of .
The heads of Google, Anthropic, Meta, OpenAI, XAI and Nvidia sign what Trump called a “morally binding” calling on them to partner with external auditors on AI risk, per these images Trump posted.
Microsoft President Backs Deployment of AI Safety Evaluators - Bloomberg News
We believe strongly in the necessity to invest into alignment. • • 1. People and businesses will only use agents that are aligned with their intent and values. If we do not build models aligned with people and businesses, then they will move to more aligned options. • • 2. Every lab should have a strong governance framework across training and deployment. This should include external evaluators, which are best practice for transparency, and independent oversight on things like safety criteria for model launches. • • 3. Every lab will need to operate within institutional protections of democratic countries. This means lab face significant liability if their models cause harm. This will push the ecosystem in the right ways. • • 4. Advances in the field are ultimately downstream of compute and resource allocation. Racing on recursive self-improvement is one of the riskiest pathways for potential loss of control to powerful models. Meta is committing the significant majority of our compute towards serving people rather than racing on racing on RSI, and over labs can choose to do the same. • • AI is a very powerful technology, and there is immense responsibility in developing it safely alongside the right checks and balances.
Mistral AI denounces Anthropic's double standards regarding the risks associated with AI
What is this event about?
On September 12, 2026, Anthropic committed to giving embedded third-party evaluators permanent, ongoing, employee-level access to its AI models and training processes.
This market will resolve to "Yes" if the listed company announces a commitment to grant one or more third-party evaluators ongoing employee-level access to its systems, AI models, and training processes during development by October 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".
Arrangements limited to evaluating models before release will not qualify. Current agreements with the Center for AI Standards and Innovation (CAISI) to provide pre-deployment access to AI models will not qualify.
A qualifying announcement must be made through the listed company's official channels or by an authorized representative acting in an official capacity. Statements that endorse third-party evaluation, describe it as under consideration, or express an intention to explore or discuss it will not qualify. A commitment made as part of a broader agreement with other companies or with a government will qualify, provided the listed company is a party to it and it otherwise meets the requirements above.
If a qualifying announcement is made within this market's timeframe, this market will resolve to "Yes" regardless of whether the commitment is binding, is later modified or withdrawn, or is ever implemented.
The primary resolution source for this market will be official information from the listed company; however, a consensus of credible reporting may also be used.