OpenAI told its staff on Wednesday that the rollout of its upcoming language model, GPT-5.6, will be staggered after a direct request from the Trump administration. Rather than a broad public launch, the company will offer a limited preview to a handful of enterprise clients. Access to the model will be granted only after the administration reviews each request on a case‑by‑case basis.
CEO Sam Altman communicated the plan during an internal Q&A, emphasizing that the company is complying with the government’s concerns about security and export controls. "We want to move forward responsibly," Altman said, adding that the preview will allow OpenAI to gather feedback while respecting the administration’s guidelines.
The Trump administration’s involvement marks a notable departure from its earlier rhetoric, which pledged a "speed wins" strategy for artificial‑intelligence development and promised to foster an American AI export program. Instead, officials have taken a more cautious stance, issuing an export‑control directive that bars foreign nationals from accessing certain advanced AI systems.
Anthropic, a rival AI firm, felt the impact of the new policy more sharply. Earlier this month, the administration gave the company an ultimatum to suspend access to its Mythos 5 and Fable 5 models for anyone who is not a U.S. citizen, including Anthropic employees abroad. The directive effectively shut down the company’s ability to market those models internationally.
OpenAI’s limited preview contrasts with Anthropic’s broader restriction. While Anthropic must halt foreign‑national use entirely, OpenAI will still ship its model to a curated set of customers, provided each request clears a government review. Industry analysts describe the arrangement as a “more favorable deal” for OpenAI, though they caution that the precedent could lead to further regulatory scrutiny of AI technologies.
Tech leaders and policymakers alike have expressed concern about the implications of such controls. Some argue that heavy‑handed export measures could hamper innovation and limit the United States’ competitive edge in the global AI race. Others contend that the safeguards are necessary to prevent misuse of powerful language models.
OpenAI’s decision also raises questions about the future of AI governance in the United States. The administration’s willingness to intervene directly in the deployment timeline of a private‑sector product suggests a willingness to balance rapid advancement with national‑security considerations. Whether this approach will become standard practice remains uncertain, but the current episode underscores the growing tension between technological progress and regulatory oversight.
For now, OpenAI’s enterprise partners will await approval before accessing GPT-5.6, while the broader AI community watches closely to see how the government’s stance evolves.
Este artículo fue escrito con la asistencia de IA.
News Factory APP - noticias agénticas para impulsar tu SEO y AEO.