Safety & SecuritySeptember 25, 2026

Google, OpenAI and Anthropic reportedly plan a safety body

The industry-run AI standards group could launch by early 2027.

Three of the biggest AI developers are planning a shared body to write safety standards for advanced AI, according to The Information. Several outlets repeated the report on September 24. We found no public statement from Google, OpenAI or Anthropic about it.

By The Policy WonkAI SAFETY STANDARDS & SELF-REGULATION

Published: September 25, 2026, 10:31 CEST
Reading time: 4 min
Evidence: 6 sources · 1 primary

This story was generated by AI systems under the editorial rules of Silicon AI News and checked against the sources it links. Found a mistake? Email editor@siliconainews.com. How we report

In brief

The Information reported on September 24 that the three companies are working on an industry-run body to set safety standards for the most powerful AI models. It is tentatively called the Standards Authority for Frontier AI. The companies have not announced it publicly.

New to this? Read it in simple words

Google, OpenAI and Anthropic are planning a group that would write safety rules for powerful AI.
The news comes from a report by The Information. The companies have not confirmed it.
The group would help outside testers check AI models before they are released.
Critics worry that companies checking themselves could favour the biggest labs.

Words to know

Standards body: A group that writes shared rules that companies agree to follow.
Frontier AI: The most advanced and powerful AI models of the moment.
Auditor: An independent expert who checks whether rules are really followed.

What the report says

According to The Information, Google, OpenAI and Anthropic are working to set up a standards body for frontier AI, the most advanced models. It is tentatively called the Standards Authority for Frontier AI, or SAFA.

It could launch by the end of this year or early next year. It would be run by the industry, not by a government.

PYMNTS, citing the report, says the body would support outside groups that test models before release. It would also set out how AI developers should report safety and security incidents.

It would define the labs’ voluntary safety commitments and set qualifications for independent auditors. A working group is also considering whether the body should test models itself.

Sources1 2 3 5

01OUTSIDE TESTINGBEFORE RELEASE

02INCIDENT REPORTSSHARED RULES

03SAFETY COMMITMENTSVOLUNTARY

04AUDITORSQUALIFICATIONS

REPORTED PLAN 01

An industry-run safety body, with four reported jobs.

Reported by The Information. The companies have not announced the plan.

Who could run it

BankInfoSecurity reports that the companies approached several people for roles in the group. They include Sriram Krishnan, a former White House AI policy adviser, and Arati Prabhakar, a former Biden administration technology official.

The list also includes former Secretary of State Condoleezza Rice and venture capitalist David Friedberg. PANews, citing the report, says Krishnan was invited to be chief executive.

SAFA would be a self-regulating body, like FINRA, which oversees US stockbrokers. It would not need approval from Congress or the president. But to have any legal power, it may need to register with a government agency. FINRA, for example, is registered with the Securities and Exchange Commission.

BankInfoSecurity notes that it is unclear whether the labs will mainly fund it.

Sources1 4 5

Why it matters, and the doubts

The idea is not new. In a July 14 essay, Google DeepMind co-founder Demis Hassabis floated a self-regulating body modelled on FINRA, BankInfoSecurity reports.

At the UN Security Council on September 23, Anthropic’s Dario Amodei also called for shared global standards for testing AI and a system for reporting AI security incidents.

The industry already has the Frontier Model Forum, founded in 2023, which Microsoft has also joined. PYMNTS reports that the Forum is in contact with the planned body.

Critics worry that a body led by the largest labs could shut out open-source developers and smaller rivals, PYMNTS reports. The plan is not final, and details may change.

Sources2 4 6

Sources

Every fact in this story comes from the sources below. Open them to check our work.

1
Primary source · September 24, 2026Google, OpenAI and Anthropic AI Safety Group Takes Shape The Information
2
Research · September 24, 2026OpenAI, Google and Anthropic Join Forces to Set AI Safety Standards PYMNTS
3
Research · September 24, 2026Google, OpenAI and Anthropic move closer to AI safety standards body Proactive Investors
4
Research · September 24, 2026Google, OpenAI, Anthropic Plan Frontier AI Standards Body BankInfoSecurity
5
Research · September 24, 2026Google, OpenAI and Anthropic’s AI safety organization is taking shape PANews
6
Research · September 23, 2026Sam Altman, Dario Amodei urge UN Security Council to adopt international AI standards CNN

How we checked this story

The Information’s report is behind a paywall, so we relied on PYMNTS, Proactive Investors, BankInfoSecurity, and PANews, which all credit it. Reports differ on some details, such as how final the plan is, so we credit each detail to its source. We found no statement from the three companies.

Story mapREPORTED PLAN

How Frontier AI Standards Body connects

Industry Self-Regulation
Frontier AI Standards Body the companies would set their own rules , connected to Industry Self-Regulation
Incident Disclosure
Frontier AI Standards Body would set rules for incident reports , connected to Incident Disclosure
Also in 9 other stories
Frontier Model Tests
Frontier AI Standards Body would support testing before release , connected to Frontier Model Tests
Also in 3 other stories
UN Council AI Meeting
Frontier AI Standards Body Amodei asked the UN for shared testing standards , connected to UN Council AI Meeting
Also in 1 other story
OpenAI Outside Testing
Frontier AI Standards Body builds on outside testing of models , connected to OpenAI Outside Testing
Also in 1 other story

The story so far

Earlier stories on the same thread, oldest first.

Safety & SecuritySep 22

Hacktron used Claude Opus 5 to reach OpenAI’s private code

The security researchers started with one image upload.

Hacktron AI says it chained a forum image bug and a sign-in flaw to take over an OpenAI employee account and open a pull request in OpenAI’s private code repository. OpenAI and Discourse fixed the flaws, and OpenAI paid a $6,500 bounty.

5 min read

Safety & SecuritySep 23

OpenAI says outside experts can test its models in training

It has not named a partner yet.

OpenAI published a framework on September 22 for independent safety checks during the training, testing, and use of its models. It lists four areas for outside review and seven principles. It names no partner, start date, or access terms, and testers do not get a full right to publish.

5 min read

Policy & LawSep 24

Altman and Amodei warned the UN Security Council of AI risks

No decision followed, and the US rejects new global AI bodies.

At a UN Security Council meeting on September 23, Yoshua Bengio, OpenAI’s Sam Altman, Anthropic’s Dario Amodei, and Hugging Face’s Clément Delangue urged countries to work together on AI risks. Amodei proposed narrow global bans and shared safety tests. No statement or resolution was adopted.

5 min read

Latest news

All news

Policy & LawOct 1

California bars firing or disciplining workers by AI alone

Newsom signed the No Robo Bosses Act with a dozen other bills on September 30. From July 2027, a human must check any discipline or firing that leans mainly on an automated system, and workers must be told in writing.

Governor Gavin Newsom signed SB 947, the “No Robo Bosses Act”, on September 30, along with 12 other bills his office grouped as AI and worker protections. From July 1, 2027, California employers may not rely solely on an automated decision system to discipline or fire workers. When such a system was the main basis, a human must corroborate the decision, and the worker must get a written notice.

4 min read

Safety & SecurityOct 1

OpenAI blames Moonshot-linked users for a copying campaign

OpenAI says the accounts tried to pull its models’ hidden reasoning in July, until it shut the campaign down. It published no evidence for the link to the Kimi maker, which had not responded publicly.

OpenAI said on September 30 that it had stopped a coordinated campaign to extract the hidden reasoning of its models. It attributes a core cluster of the activity to people associated with Moonshot AI, the Chinese developer of Kimi. The campaign peaked on July 24 and 25 with 16,000 attempts from more than 4,000 users, and OpenAI says it was fully disrupted by July 28.

3 min read

Safety & SecurityOct 1

DeepMind watermarks AI-designed proteins for biosecurity

SynthID Bio hides a signature in a protein’s building blocks, and lab tests found marked designs worked as well as unmarked ones. It could help DNA makers tell designs from trusted AI tools apart from the rest.

Google DeepMind introduced SynthID Bio on September 30, a way to watermark proteins designed by AI without breaking them. In lab tests on three targets, watermarked protein binders matched unwatermarked ones on how often they worked, how tightly they bound and how varied they were, DeepMind says. It is sharing the code, lab data and model weights with researchers.

4 min read

Policy & LawOct 1

Connecticut’s AI rules on layoffs and whistleblowers start

From October 1, employers must say whether AI played a part in mass layoffs, and makers of the biggest AI models may not punish staff who warn of catastrophic risks. AI subscriptions and AI-made media get new rules too.

Large parts of Connecticut’s new AI law took effect on October 1. Employers that file mass-layoff notices must now tell the state whether the layoffs relate to their use of AI. Developers of the biggest AI models may not retaliate against staff who report catastrophic risks, and AI subscription services must spell out their limits before customers pay.

4 min read

BusinessOct 1

ElevenLabs is now valued at $22 billion after a share sale

Employees sold $300 million of stock at twice February’s valuation, and the company raised no new money. ElevenLabs says businesses now bring in 55% of its revenue, and voice agents are its fastest-growing product.

ElevenLabs, the AI voice company, said on September 30 that a $300 million share sale by employees values it at $22 billion. Wellington and T. Rowe Price led the deal, and investors including EQT, Goldman Sachs and Singapore’s GIC bought in for the first time. The valuation is double the $11 billion of its funding round in February.

3 min read