OpenAI, Anthropic and Google have confirmed they are working together to create a voluntary AI industry standards body, with representatives from the three companies meeting regularly since July to hash out a proposal. OpenAI’s global policy chief Chris Lehane said this week that the engagement had been under way for several weeks, escalating industry efforts to respond to growing concern that advanced AI poses economic and security risks.
The idea originated from a proposal by Google DeepMind chief Demis Hassabis for a self-regulatory body modelled on FINRA, the Financial Industry Regulatory Authority that oversees US brokers and investment firms. Hassabis had called in July for an organisation to test the most powerful AI systems before public release. The talks gained further momentum after Anthropic CEO Dario Amodei publicly urged AI companies to cooperate on auditing and testing frontier models.
Why Are OpenAI, Anthropic and Google Building a Standards Body Now?
With calls to slow AI development gaining traction among lawmakers and civil society, and talks inside the US government reportedly at a standstill, the three labs are moving to establish a voluntary body of their own rather than wait for government-mandated regulation. OpenAI CEO Sam Altman has said he favours a testing and auditing organisation for the industry but believes major AI labs will need to set it up without US government backing, given the current political gridlock on AI legislation.
What Does This Mean for AI Adoption in Indian Industry?
For Indian enterprises increasingly embedding OpenAI, Anthropic and Google models into products, a shared industry standards body could eventually mean more consistent safety benchmarks, audit trails and testing disclosures across vendors, reducing the compliance guesswork Indian CIOs currently face when choosing between competing foundation model providers. It also matters for India’s own AI governance conversation: the IndiaAI Mission and MeitY have been developing domestic frameworks, and a credible global self-regulatory body could shape how Indian regulators eventually benchmark foreign AI systems entering the market, from enterprise copilots to consumer chatbots.
Industry Reaction and Expert Commentary
The three-way collaboration is notable given how fiercely OpenAI, Anthropic and Google compete on model performance and enterprise contracts. Analysts see the move as an attempt to get ahead of fragmented, country-by-country AI regulation by presenting a unified, self-policed alternative. Skeptics note that self-regulatory bodies in other industries have had mixed records on independence, and that any standards body’s credibility will depend heavily on whether it includes external auditors rather than being run solely by the labs it is meant to oversee.
What Happens Next?
No formal launch date or governance structure for the proposed standards body has been announced yet, and discussions remain at the proposal stage among the three companies. Watch for whether other major labs, including Meta and Microsoft, are invited to join the initiative, and whether the body’s first public output is a testing framework, a disclosure standard, or simply a founding charter.
Frequently Asked Questions
Which companies are working on the AI standards body?
OpenAI, Anthropic and Google (via Google DeepMind) have confirmed they are in talks to create a voluntary AI industry standards body, with discussions ongoing since July 2026.
What model is the proposed AI standards body based on?
The idea, proposed by Google DeepMind CEO Demis Hassabis, is modelled on FINRA, the self-regulatory body that oversees US brokers and investment firms.
Why are the companies pursuing self-regulation instead of waiting for government rules?
With government-led AI regulation talks at a standstill in the US and public pressure to slow AI development growing, the labs are moving to set up their own voluntary oversight body rather than wait for legislation.
Leave a comment