OpenAI and Anthropic Push Washington to Keep Up with AI

The Companies’ Endorsement Shifts the AI Safety Debate

The endorsement by OpenAI and Anthropic has shifted the AI safety debate from a general concern to a more concrete policy challenge: who decides when progress is too fast, and what can governments actually do about it? This move presents Washington with a problem that is easier to state than to administer. Specifically, the question arises of whether the government should attempt to regulate the pace of AI development.

According to a report by The Washington Post, the rival labs supported a letter signed by hundreds of AI company employees, urging U.S. officials to prepare ways to temper AI progress if it accelerates too quickly. This includes diplomatic efforts with other countries. While this does not provide clear answers on how a limit would be measured, enforced, or reconciled with the companies' own competition, it does shift the argument from whether AI safety should be taken seriously to what kind of public authority could act before the market sets the pace on its own.

The Central Question: Speed

The endorsement is not framed as a call to halt AI development immediately. Instead, it focuses on a conditional point: if progress becomes too rapid, governments should have a way to slow it down. This distinction is crucial. A standing pause would be one form of policy, while a pacing system would require triggers, evidence, and a decision-maker capable of determining when development or deployment has crossed a line.

OpenAI and Anthropic are not outside observers of this system. They are major players in the frontier AI race, which makes their support both notable and complex. Any rule strong enough to affect the industry would likely impact them as well.

Employees Pressing from Inside the Industry

The letter's support base came from hundreds of workers at AI companies, according to The Washington Post. Their appeal reflects concerns from individuals close to the technology rather than just academics, activists, or regulators.

This concern is partly about incentives. Companies compete for customers, talent, infrastructure, and investor confidence. Faster releases can offer an advantage, while slower reviews can create time for testing, outside scrutiny, and public debate. This tension underlies the concept of "pacing." It suggests that the release rhythm for powerful systems should not be set solely by competitive pressure. However, the unresolved issue remains what proof would justify slowing that rhythm and who would be trusted to make the call.

Washington Already Has a Safety Channel

The federal government is not starting from scratch. In 2024, the National Institute of Standards and Technology (NIST) announced agreements with OpenAI and Anthropic for AI safety research, testing, and evaluation through the U.S. AI Safety Institute. These agreements created a framework for the institute to access major new models from each company before and after public release. The work aimed to evaluate capabilities, study safety risks, and explore ways to reduce those risks.

Elizabeth Kelly, then director of the U.S. AI Safety Institute, emphasized that safety was essential to technological innovation and called the agreements an important milestone. The institute also planned to coordinate with the U.K. AI Safety Institute. These arrangements were framed as technical collaboration, but the new endorsement raises a separate policy question: whether voluntary safety cooperation could evolve into something closer to government influence over the pace of AI deployment.

The Approval Line Remains Sensitive

One reason the debate is challenging is that slowing AI can sound like requiring permission before new systems are released. That is a line some industry leaders have approached carefully. Reuters reported in June that OpenAI CEO Sam Altman was expected to urge U.S. lawmakers not to require AI developers to obtain government approval before releasing new models. This position does not rule out all oversight but highlights the difference between safety review and pre-release licensing.

There is also a competition concern. Large companies often handle compliance systems, reporting duties, evaluations, and security requirements more easily than smaller rivals. Critics may question whether new rules would protect the public, strengthen incumbents, or do both. This skepticism does not negate the safety argument, but it means any pacing mechanism would need to be judged not only by its stated goal but also by who gains power from its design.

A U.S.-Only Answer May Be Too Weak

The letter’s call for international diplomacy points to a practical challenge: AI development does not stop at national borders. Labs, researchers, cloud infrastructure, chips, and customers are part of a global system. If the United States slowed only domestic companies, critics could argue that foreign competitors would benefit. If the U.S. sought broader limits, governments would need shared definitions, verification methods, and enforcement tools.

This is far more complex than announcing a principle. AI involves software, compute, data, research practice, and commercial deployment simultaneously. A workable framework would have to account for private labs, open-source models, academic work, and companies operating across jurisdictions.

The Policy Design Is Still Missing

The endorsement gives policymakers a clearer mandate to examine AI pacing, but it does not provide the machinery to implement it. Possible approaches could include mandatory safety evaluations, compute thresholds, incident reporting, coordinated release delays, international standards, or emergency powers tied to specific model capabilities.

Each option comes with trade-offs. A light system may not change company behavior, while a heavy system could slow useful products, concentrate power among large firms, or give agencies authority they are not ready to use. The immediate takeaway is not that an AI speed limit now exists. Rather, it is that two leading AI companies have backed a call for the government to consider when one might be needed, leaving Washington with the hardest questions still unanswered.

Share :