U.S. Discusses Voluntary Standards for AI Model Releases With AI Companies

The US Government Discusses Voluntary Standards for AI Model Releases with AI Companies

The US government is negotiating with leading AI companies about voluntary standards for releasing new models — according to Financial Times, an official announcement could come soon.

What the Parties are Negotiating

According to Financial Times, the discussed standards should establish benchmarks and timelines for advanced AI models, as well as clarify who receives access to these models — both within the United States and abroad.

• Negotiations are taking place between the US government and AI companies
• The standards are voluntary — not legislatively mandatory
• Standards concern the procedure for releasing new models
• Access to models both within the country and abroad is being discussed
• The source of information about the negotiations is Financial Times

The very fact that such negotiations are occurring shows that the federal government is seeking an alternative to both complete non-intervention in AI development and strict legislative regulation modeled after individual states. Financial Times does not reveal which specific companies are participating in the negotiations, but given the scope of the issues being discussed — from benchmarks to controlling international access — it is clear that the largest developers of advanced models are involved.

Why a Voluntary Format Was Chosen

The voluntary obligation format allows the government and AI companies to agree on the rules of the game without going through a lengthy legislative process. For companies, this means predictability: pre-established benchmarks and timelines instead of fragmented requirements that may emerge at the state level.

For AI laboratories, voluntary standards reduce the risk of facing multiple parallel and sometimes contradictory requirements from several states simultaneously. However, the voluntary nature of the agreement means that federal authorities may not have direct enforcement mechanisms at this stage — companies either adopt the standards or remain outside the framework of the agreement.

This also aligns with the general approach of federal authorities to technological regulation in recent years: Washington has systematically preferred targeted agreements with industry to centralized lawmaking, arguing that rigid federal standards simply cannot keep pace with the speed of technological development.

What Model Access Rules Mean

A separate part of the negotiations concerns who receives access to new advanced models — both within the United States and beyond. This directly relates to national security issues and controlling the spread of the most powerful AI systems among foreign players, including potential geopolitical competitors.

Such a formulation of standards transforms the negotiations not only into a product safety issue but also into a foreign policy instrument — access to cutting-edge models becomes a resource that the US government can regulate alongside technology and equipment exports. For US allies, this potentially means more predictable access to cutting-edge models, while for countries under export restrictions, the negotiations continue the current course of technological protectionism.

What This Means

If the negotiations conclude with an official announcement, the United States will gain its first notable benchmark in the spirit of industry self-regulation — a compromise between the desire to control the release of increasingly powerful models and the reluctance to slow down competitive dynamics through rigid legislation. For the rest of the world, this will signal which path the United States ultimately took: soft law instead of direct regulation. Until the official announcement, details remain subject to negotiation, which means the final set of requirements may differ from what is currently being discussed behind closed doors between Washington and the world's largest AI laboratories.

Financial Times promises to monitor the development of negotiations and update information as new details about the content of future standards emerge.