Google, OpenAI, Anthropic Form Joint AI Safety Regulator

What Happened

Google, OpenAI, and Anthropic announced plans this week to form a joint industry body modeled on FINRA, the Financial Industry Regulatory Authority, to establish safety rules for powerful artificial intelligence models. The announcement came one week after the Trump administration declined to create a federal AI regulator.

Background

The three companies are the developers of the most widely deployed large-scale AI systems currently in commercial use. OpenAI produces the GPT series of models and the ChatGPT platform. Google develops the Gemini model family across its consumer and enterprise products. Anthropic, founded in 2021 by former OpenAI researchers including CEO Dario Amodei, develops the Claude series of models.

FINRA, the model cited for the proposed body, is a non-governmental organization authorized by Congress to regulate broker-dealers in the United States. It sets conduct standards, enforces rules, and operates independently of direct government control while remaining subject to oversight by the Securities and Exchange Commission. The comparison indicates the companies are proposing a self-regulatory organization with binding or quasi-binding authority over member firms, rather than a voluntary standards body.

The announcement follows a period of active but unresolved debate in Washington over how to govern advanced AI systems. The Trump administration passed on establishing a dedicated federal AI regulator, a decision reported in the days immediately before this announcement. House Speaker Mike Johnson confirmed that a White House meeting between President Trump and technology company executives, scheduled for Tuesday, would focus on finding a balance between innovation and oversight.

President Trump separately confirmed a dinner meeting with Anthropic CEO Dario Amodei, while reiterating his position dismissing AI safety concerns as overstated.

What the Proposal Involves

The three companies have not released full structural details of the proposed body. Based on reporting from Forbes, the organization is intended to set safety rules specifically for powerful AI models, a category that in regulatory and policy contexts typically refers to frontier or large-scale foundation models capable of general-purpose reasoning and autonomous task completion.

The FINRA comparison, used by the companies or their representatives in describing the initiative, points toward a model in which member organizations fund and operate the regulatory body, set industry-wide standards, and enforce compliance among members. FINRA's structure gives it authority to fine, suspend, or bar member firms and individuals, though its jurisdiction is bounded by its Congressional authorization.

Whether the proposed AI body would have comparable enforcement mechanisms, and which companies beyond the founding three would be eligible or required to join, had not been specified in available reporting at the time of publication.

Broader Context

The announcement coincides with warnings from AI researchers about the pace of development. A Reuters report published September 29, citing current and former researchers at OpenAI and Google DeepMind, stated that researchers are cautioning companies are doing too little to address risks from self-improving AI systems. Microsoft co-founder Bill Gates, in a separately reported interview, called for safeguards beyond industry self-regulation, though that story has been previously covered in this wire service's recent reporting.

Just Security, a legal and policy publication, published analysis this week calling for a formal, multistakeholder, internationally relevant AI accountability mechanism, citing what it described as an existing accountability gap in AI governance.

A separate op-ed published by news.lavx.hu argued that Congress should investigate frontier AI laboratories, contending that dramatic safety claims from OpenAI and Anthropic have shaped public debate in ways that merit legislative scrutiny.

What Happens Next

The White House meeting between President Trump, House Speaker Mike Johnson, and technology company executives is scheduled for Tuesday, where AI oversight is expected to be among the topics discussed.