Trump administration asks OpenAI to delay new model release, The Information reports

Reuters | June 25, 2026 at 08:55 PM UTC
Bearish 79% Confidence Majority Agreement
Read Original Article

Key Points

  • The model, GPT 5.6, will be released in a restricted preview format to a small group of partners rather than a full public launch
  • Two government agencies drove the request: the Office of the National Cyber Director and the Office of Science and Technology Policy
  • The government will maintain approval authority over each customer's access during the preview period, representing unprecedented oversight of an AI model release

AI Summary

Summary

The Trump administration has requested OpenAI delay the full release of its new AI model, GPT 5.6, due to security concerns, according to a report by The Information citing an internal company memo.

Key Details:

OpenAI CEO Sam Altman informed staff during a Q&A session that the company will proceed with a limited preview release, making the model available only to a small group of partners initially. The government will implement a customer-by-customer approval process during this preview period, representing an unprecedented level of federal oversight for an AI model launch.

Government Involvement:

The request for a staggered release emerged from discussions with two key federal agencies:

  • Office of the National Cyber Director
  • Office of Science and Technology Policy

Market Implications:

This development signals increasing government scrutiny of advanced AI systems and potential security risks associated with powerful language models. The intervention establishes a precedent for federal involvement in private AI company product launches, which could impact:

  • Future AI development timelines across the industry
  • Competitive dynamics among AI companies
  • Investor expectations regarding AI product rollouts
  • Regulatory framework for emerging AI technologies

The controlled release approach may delay OpenAI's ability to monetize GPT 5.6 commercially and could provide competitors additional time to close technology gaps. This also raises questions about how similar restrictions might apply to other major AI players like Google, Microsoft, and Anthropic.

The news was reported on June 25, with no specific timeline provided for when broader public access might be granted.

Model Analysis Breakdown

Model Sentiment Confidence
GPT-5-mini Bearish 75%
Claude 4.5 Haiku Bearish 78%
Gemini 2.5 Flash Neutral 85%
Consensus Bearish 79%