OpenAI has asked members of Congress whether rival artificial intelligence companies could legally agree to slow development, according to WIRED. The outreach comes as OpenAI Chief Scientist Jakub Pachocki urges the industry to establish shared safety standards and consider voluntary pauses until developers can better demonstrate that their systems are safe.

WIRED, citing people close to OpenAI, reported that the company asked lawmakers in recent weeks whether an agreement among AI developers to slow progress could violate antitrust laws. An arrangement that restricted competition could raise antitrust concerns, although legislation introduced by Sens. Adam Schiff and Jim Banks would protect certain security-related collaborations if the participants provide advance notice to the Justice Department.

Experts say funding and competitive pressures make it difficult for companies to slow development independently. “Commercial and geopolitical competition in the AI space is incredibly intense, leading to a concerning dynamic where companies are incentivized to release products before their risks are fully understood,” Miranda Bogen, chief technologist at the Center for Democracy and Technology, told Decrypt.

“Even when internal staff knows more research and testing is needed, their companies are facing immense pressure to cut corners and skip critical safety tests, despite evidence piling up about the consequences of moving too fast,” she added.

The discussion is unfolding as the United States and China compete for leadership in AI. U.S. President Donald Trump delayed an AI executive order in May over concerns that it could weaken America’s lead over China. He later signed the order in June, creating a voluntary review process for advanced models before their release.

Financial incentives may pose an even more direct obstacle to coordination. “Every advance under current conditions yields many millions or billions more in funding and puts the creators of that advance in a greater position of power and influence,” Duncan Sabien, head of communications at the Machine Intelligence Research Institute, told Decrypt.

Sabien said the arms-race dynamic is real and that OpenAI and Anthropic face pressure to continue advancing despite uncertain risks. Each new system can help produce more capable successors, creating the possibility that a rival will pull ahead if either company slows down without a shared agreement.

“This is especially true since gains in intelligence are compounding; each new system makes it easier to train and deploy the next, more-advanced system,” he said. “Sans some sort of coordination mechanism, stepping back just means the other guy gets a lead.”

OpenAI paused internal Astra work in August because it lacked stronger safeguards addressing cybersecurity concerns. In February, however, both OpenAI and Anthropic relaxed their safety commitments. Anthropic Chief Science Officer Jared Kaplan argued that slowing development independently made less sense while competitors continued to advance.

The debate intensified this week after former Anthropic engineer Jacob Coxon publicly announced his resignation, citing concerns about the risks that continued AI development poses to humanity.

“The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon said on X, adding that it was not a “marketing stunt.” “If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible—but I hear the same people express fear privately. No other human activity poses this level of danger,” he wrote.

Sabien said researchers could accomplish little by leaving if someone else simply replaced them. “If enough of them achieve common knowledge that they should all stop, then they can break through the coordination barrier and stop together rather than just being replaced by the next slightly-less-cautious person,” he said.