Jakarta, ThedailyID — OpenAI, Google and Anthropic have reportedly discussed creating an industry standards body for artificial intelligence (AI).
The proposed organization would develop standards for testing and deploying advanced AI models. Discussions began before several AI safety developments emerged in recent weeks.
According to people familiar with the talks, the idea followed a July essay by Google DeepMind co-founder Demis Hassabis.
Hassabis proposed a US-led organization modeled after the Financial Industry Regulatory Authority (FINRA). The body would test advanced AI models before companies release them or deploy them widely.
He suggested a public-private partnership overseen by the government but funded by the industry. Independent technical experts and members of the open-source community would also participate.
The discussions remain ongoing. One source said they could continue with or without involvement from US President Donald Trump’s administration.
Meanwhile, not every technology leader supports the proposal. Meta CEO Mark Zuckerberg reportedly advised Trump against backing the plan during a phone call this summer.
Representatives for Google, OpenAI and Anthropic declined to comment or did not respond to requests for comment.
Currently, the industry has few common standards governing how companies test advanced AI models. Concerns have grown after several incidents involving AI agents during testing this summer.
In one serious case, an OpenAI agent reportedly escaped its testing environment and hacked another company’s systems to cheat during a cybersecurity test.
The White House also has a voluntary system that allows AI companies to submit new models for government review up to 30 days before public release. However, officials have not published the eligibility requirements or details of the review process.
OpenAI has indicated that it supports stronger safety standards. Chief Scientist Jakub Pachocki said shared safety standards and international coordination should become a priority.
“I believe shared safety standards and international coordination in further AI development need to be a priority now,” Pachocki said earlier this month, according to CNN, Wednesday, September 16.
Pachocki also said OpenAI had spoken with external organizations about potential concrete standards. The company expects to share more developments in the coming months.
OpenAI CEO Sam Altman has also expressed support for working with other companies to develop industry standards. However, the companies have yet to agree on the structure or scope of such a system.
US House Speaker Mike Johnson said competing AI companies and policymakers still lack consensus on appropriate rules and limits.
“This has to be a partnership with the industry itself, with the companies developing this technology, as well as policymakers and legislators,” Johnson said.
