Archive · 16 Sep 2026
AI Firms Propose Safety Standards Body Amid Growing Fears
OpenAI, Anthropic and Google are developing a proposed AI standards body modeled on FINRA.
The plan was proposed by Google DeepMind chief Demis Hassabis and confirmed by OpenAI on Tuesday.
Former AI employees and researchers have warned that rapid development could create serious risks to humanity.
Cohere CEO Aiden Gomez criticized the proposal, arguing that major US AI companies could control who writes the rules.
Donald Trump and Nvidia CEO Jensen Huang have dismissed or criticized predictions that AI could destroy humanity.
- Who
- OpenAI, Anthropic and Google are working on the proposal, with the idea attributed to Google DeepMind chief Demis Hassabis.
- What
- The companies are considering a new AI industry standards body modeled on FINRA.
- When
- OpenAI confirmed the plan on Tuesday; related warnings and comments were made in the preceding days and weeks.
- Where
- The proposal concerns the global AI industry; several comments were made in the United States, including at a conference in Los Angeles.
- Why
- The initiative is intended to address growing concerns about the safety risks of increasingly powerful AI systems and the pace of their development.
Safety advocates
Industry critics and skeptics
Whether AI development should slow down
Safety advocates
Former employees and researchers, including Jacob Coxon, Bilal Chughtai and Anthropic CEO Dario Amodei, warned that AI development may be moving too quickly and could pose severe risks.
Industry critics and skeptics
Donald Trump and Nvidia CEO Jensen Huang rejected or criticized extreme predictions about AI destroying humanity; Huang said such claims are not grounded in science.
Who should write AI safety rules
Safety advocates
OpenAI, Anthropic and Google are pursuing common industry standards to address safety concerns as AI systems become more powerful.
Industry critics and skeptics
Cohere CEO Aiden Gomez argued that the key dispute is who participates in writing the rules and whose interests they protect, accusing major US AI companies of forming a cartel.
- Proposed body
- A standards organization for the artificial intelligence industry
- Companies involved
- OpenAI, Anthropic and Google
- Model
- The Financial Industry Regulatory Authority, or FINRA
- Proposal credited to
- Demis Hassabis, chief of Google DeepMind
- Main concern
- Potential dangers from rapidly developing, increasingly powerful AI systems
- Industry criticism
- Cohere CEO Aiden Gomez accused major US AI companies of forming a cartel under the cover of AI safety
- Public disagreement
- Donald Trump dismissed danger warnings as hoaxes, while Jensen Huang called predictions of AI destroying humanity unscientific and irresponsible