US government is now a customer gatekeeper for OpenAI Sol and Claude Mythos 5

On 2026-06-26, two related things happened at the US frontier-AI frontier. OpenAI began a limited preview of GPT-5.6 Sol with the customer list “shared with the government” (OpenAI, 2026-06-26). On the same day, the Commerce Department lifted its two-week export block on Anthropic’s Claude Mythos 5, releasing it to “more than 100 US institutions, including major companies and government agencies” via a letter from Commerce Secretary Howard Lutnick (Semafor, 2026-06-26, 6:30pm EDT). Together the two actions establish a new distribution model: each new customer of a US frontier lab’s most capable model is, in practice, approved by the Trump administration’s Commerce Department. Semafor frames this as “the framework for overseeing AI is being built on the fly”; the Washington Post calls it “expanding its recent policy of vetting companies that want access to the latest artificial intelligence technology.” No Federal Register entry, no executive order, no published criteria. The policy is the story. The model releases are evidence.

What happened — the regime

OpenAI’s announcement describes the GPT-5.6 Sol preview as “a limited preview for a small group of trusted partners whose participation has been shared with the government.” The Lutnick letter grants a license exemption for “entities identified in Annex A to this letter and their foreign national employees.” The Annex A list is not public. Per WaPo, “The Trump administration is requiring both Anthropic and OpenAI to get approval for each new customer of their most powerful AI technology.” The word “both” is doing work: this is no longer a one-off Mythos case but a pre-approval regime across the two US frontier labs.

No public rules. Semafor: “The framework for overseeing AI is being built on the fly, and many users of the powerful tools … remain in the dark.” OpenAI names a “cyber Executive Order framework” as the path out, but no such order has been published as of 2026-06-27.

OpenAI pushes back. “We don’t believe this kind of government access process should become the long-term default. It keeps the best tools from users, developers, enterprises, cyber defenders, and global partners who need them. We are taking this short-term step because we believe it is the strongest path to broader availability in the coming weeks.” Sam Altman on X: “I just don’t like the idea of the government picking the customers.” Per Semafor, “European officials and other US allies have expressed frustration.” Britain’s AI Security Institute is the only non-U.S. entity with access to GPT-5.6.

What happened — the model releases

GPT-5.6 series, new naming scheme. Per OpenAI: “Sol, our flagship model; Terra, a balanced model for everyday work; and Luna, a fast and affordable model. Terra has competitive performance to GPT-5.5 while being 2x cheaper and Luna brings strong capability at our lowest cost.” The number identifies generation; Sol, Terra, Luna identify durable capability tiers.

Pricing per 1M tokens. Sol: $5 input / $30 output. Terra: $2.50 / $15. Luna: $1 / $6. On caching: “cache writes are billed at 1.25x the model’s uncached input rate, while cache reads continue to receive the 90% cached-input discount.” A new max reasoning effort and an ultra mode leveraging subagents ship with Sol.

Capability claims — all vendor-attributed. Terminal-Bench 2.1 SOTA. GeneBench v1 “achieves stronger results than GPT-5.5 while using fewer tokens.” ExploitBench²: “GPT-5.6 Sol is competitive with Mythos Preview using only ~1/3 of the output tokens.” On the Cyber Critical threshold: “GPT-5.6 Sol does not cross the Cyber Critical threshold under our Preparedness Framework … it did not autonomously produce a functional full-chain exploit under the conditions tested.” All are OpenAI’s own measurements.

The Mythos 5 letter. Lutnick to Anthropic: “I have determined that appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model.” Semafor: “The letter is silent on Fable 5 … People close to the talks said they are moving toward releasing Fable as well, though that timeline is unclear.”

Why it matters now

1. It is a structural shift, not an event. Two frontier labs are now under the same Commerce Department customer-vetting regime, with the same day as the visible evidence.
2. The regime is being built without public rules. “Built on the fly” is Semafor’s framing. Rep. Sam Liccardo to WaPo: “the unreviewable gatekeeper of the world’s most powerful technology, and nobody inside or outside the administration can say what standards it’s using.”
3. OpenAI is publicly unhappy, in its own announcement. The most useful single line for a builder audience.
4. Foreign users, including US allies, are now dependent on US government decisions. The British AI Security Institute is the only non-U.S. entity with access to GPT-5.6.

Practical implications

• Frontier-model procurement is now a regulatory event, not a sales event. Until the cyber Executive Order lands, the US government is the de facto customer-vetting step for OpenAI Sol and Anthropic Mythos 5.
• GPT-5.6 Terra and Luna are NOT under the customer-vetting regime, per OpenAI — the vetting is for the Sol preview.
• Cache economics change for GPT-5.6. Cache writes at 1.25x uncached input; 30-minute minimum cache life; explicit cache breakpoints.
• Do not plan production workloads on the GPT-5.6 Sol preview. Broader availability is “in the coming weeks,” not a calendar date.
• For non-US and non-Annex-A companies, the Mythos 5 question is the regulatory event of the week. Per WaPo, “only U.S.-based companies have access.”

Risks and caveats

1. All GPT-5.6 capability claims are OpenAI’s. The article attributes; it does not assert.
2. “Does not cross the Cyber Critical threshold” is OpenAI’s self-assessment under OpenAI’s own Preparedness Framework.
3. The “new regulatory regime” is Semafor’s framing, confirmed by WaPo and OpenAI. No more specific legal characterisation.
4. No published rule, no executive order, no Federal Register entry. The article does not infer a process not in one of the three primary sources.
5. The Annex A customer list for Mythos 5 is not public. The article does not name specific customers.
6. The Fable 5 status is unresolved. Semafor: “moving toward releasing Fable as well, though that timeline is unclear.”

What to watch

1. The cyber Executive Order itself — OpenAI names it as the path out.
2. Fable 5 timeline (Semafor: “moving toward releasing Fable as well, though that timeline is unclear”).
3. Broad GPT-5.6 availability (Sol on Cerebras “at up to 750 tokens per second in July” is a separate track).
4. Whether xAI, Google, or Meta are brought into the regime.
5. Independent benchmarks of GPT-5.6 Sol.
6. The “framework being built on the fly” materialising into a published rule.

Sources

• OpenAI — “Previewing GPT-5.6 Sol: a next-generation model” (2026-06-26)
• Semafor — “Exclusive / US releases powerful Anthropic model Mythos to some US companies” (Albergotti, Smith, 2026-06-26)
• Washington Post — “The U.S. government will decide who gets to use the latest American AI technology” (De Vynck, Arnsdorf, 2026-06-26)
• Semafor — original 2026-06-13 exclusive on the Mythos export-control trigger
• Wired — “SK Telecom / Anthropic Mythos export controls” (2026-06-13)
• OpenAI — “Updating our Preparedness Framework” (2026-06-26)