OpenAI CEO Sam Altman says his industry is already like a fictional world.
“Evidence is emerging of AI’s economic value, its importance to national security, and its acceleration of scientific discovery,” Altman said in an op-ed for the Financial Times. “In the next year or two, we expect to build systems with incredible capabilities that can provide tremendous value to the world.”
Altman said AI systems that would recently be called “science fiction” are already in place and will continue to change the world.
“Artificial intelligence will reshape the material conditions of human life on a scale that no technology has achieved since the use of electricity, and perhaps even beyond,” he wrote.
Altman’s words come at a critical time for this nascent technology, a turning point where the OpenAI CEO said a global AI framework is needed.
Altman’s proposal follows recent conversations with world leaders and tech executives at the G7 meeting in France to create a “US-led international forum.”
The organization, which could include government representatives and independent technical experts, would establish standards for the AI industry and “make that technology available to countries and companies that participate and play by the rules,” Altman said.
“It can also serve as a governance mechanism for the institute, protecting it from commercial pressures that could lead to dangerous races,” he wrote.
Most importantly, Altman stressed that unless the world comes together to set comprehensive standards, individual countries will have too much influence.
“Such international cooperation seems like a reasonable way to avoid undue concentration of power and ensure that the benefits of AI are democratized,” he wrote.
Altman did not explicitly mention the U.S. government’s request to limit the release of GPT-5.6 Sol, Terra, and Luna to a group of “trusted partners” whose names were initially shared with the government, to which OpenAI agreed.
But in a statement announcing the decision, OpenAI said such limited releases should not “become the long-term default.” As OpenAI approaches an IPO, the Trump administration’s demands also loom large. The request comes on the heels of an even more heated standoff between the Commerce Department and Anthropic, which led the Trump administration to temporarily impose export controls on OpenAI’s rivals.
Altman said the AI industry can look to established regulatory bodies for inspiration. Mr. Altman singled out the International Atomic Energy Agency, an organization under the United Nations that monitors global nuclear energy development and ensures that countries are not seeking to acquire nuclear weapons.
“This cooperation was possible even during a period of great international turmoil; the IAEA, for example, was founded at the beginning of the Cold War,” Altman wrote.
Anthropic CEO Dario Amodei, whose relationship with his former colleague Altman is notoriously controversial, called for AI regulations to mirror those of the Federal Aviation Administration in a recent essay. Mr. Amodei went further than Mr. Altman, making clear that the federal government has the power to halt the release of AI models if third-party testing reveals an “unacceptable risk.”
“There may come a time, perhaps relatively soon, when the most powerful AI systems will have to move beyond planes and cars to something akin to weaponizable nuclear material, a threat to humanity rather than ‘just’ a threat to public safety,” Amodei wrote. “If that happens, we may need more aggressive regulatory action than what I have proposed.”