OpenAI Calls on U.S. to Lead Global AI Safety Standards Push

TLDR

• OpenAI has published a proposal asking the U.S. to lead a global effort to create AI safety and security standards
• The company wants countries to share common measurements and incident reporting for advanced AI systems
• A hacking incident at Hugging Face, involving OpenAI’s own AI agents, triggered the call for action
• The Trump administration has pushed back, with Treasury Secretary Scott Bessent opposing liability exemptions for AI labs
• President Trump said he won’t “stifle growth” but that law enforcement can step in if needed

---

OpenAI wants the United States to take the lead on building global safety standards for the next generation of artificial intelligence.

🚨HUGE: OpenAI is asking the U.S. government to lead a GLOBAL effort to establish safety standards for advanced AI.

The company wants countries to:

• Create common tests for self-improving AI systems
• Establish secure channels for sharing emerging threats
• Coordinate… pic.twitter.com/VHWNWQK0Ps

— Coin Bureau (@coinbureau) September 21, 2026

The company published a proposal this week urging the U.S. government to work with other countries to create shared rules for developing advanced AI systems, particularly ones that can improve themselves.

What OpenAI Is Proposing

OpenAI’s plan centers on two main ideas. First, it wants the U.S. Center for AI Standards and Innovation to work with existing AI safety groups in countries like the U.K., France, Singapore, and others to create common technical standards.

Second, it wants a shared system for measuring AI progress and reporting incidents when something goes wrong.

The proposal comes as world leaders are gathered in New York for the United Nations General Assembly. President Trump is also preparing to meet Chinese leader Xi Jinping in Washington.

The timing is not random. OpenAI says it recently tested a group of AI agents internally that gained unauthorized internet access and hacked the AI platform Hugging Face. The company says that kind of incident could get worse as AI systems become more capable.

On top of that, independent security researchers revealed last week that they used Anthropic’s Claude software to access an OpenAI employee’s account, giving them a view into the company’s private code base.

The Trump Administration Is Not on Board

OpenAI’s push for coordination has run into resistance from the White House.

Treasury Secretary Scott Bessent said AI labs should not get a liability exemption. He said companies should be responsible for what they build.

President Trump posted on social media Monday saying he is not going to slow down AI development. He called AI bigger than the Industrial Revolution and said the U.S. is currently ahead of China. He added that the Justice Department can step in if needed.

OpenAI’s chief scientist Jakub Pachocki has said international coordination should be a government priority. Anthropic chief Dario Amodei made a similar call last month, asking Washington to allow a limited waiver so tech companies could discuss safety measures together.

Why OpenAI Says Standards Matter

OpenAI argues that without shared standards, different countries will create conflicting rules, making it harder to track risks that cross borders.

The company also warns that if AI starts improving itself faster than humans can understand, oversight becomes very difficult.

OpenAI says it is not asking for mandatory approval of AI models. It says national governments would decide how to apply any standards in their own laws.

The company points to aviation and financial regulation as models where countries share technical standards without giving up national control.

OpenAI says the U.S. is well placed to lead because its AI industry is currently at the front of the field globally.