White House Urges OpenAI to Delay Release of New AI Model Amid Safety Concerns

The White House has reportedly asked OpenAI to slow the release of its next-generation artificial intelligence model, citing concerns over national security and the potential risks associated with increasingly powerful AI systems. The request reflects the US government’s growing involvement in overseeing the development and deployment of advanced AI technologies as they become more capable and widely adopted.

According to reports, the administration has encouraged OpenAI to adopt a phased rollout rather than launching the model to the general public immediately. Under the proposed approach, access would initially be restricted to a limited number of trusted enterprise customers, researchers and partner organisations, allowing additional time to evaluate the system’s capabilities and identify any unforeseen safety issues.

The discussions involve senior officials from several government agencies responsible for science, technology and cybersecurity. Authorities are particularly focused on assessing whether advanced AI models could be exploited for cyberattacks, the creation of sophisticated misinformation campaigns, or other activities that may threaten national security and public safety. Officials believe a gradual deployment would provide valuable opportunities to monitor the technology’s real-world performance before expanding public access.

OpenAI has reportedly agreed to cooperate with the administration’s request by delaying the full-scale release of the model while additional safety testing is conducted. The company is expected to continue refining safeguards and evaluating potential risks before making the technology more broadly available. The phased rollout is also intended to help developers gather feedback from early users and address any issues that emerge during limited deployment.

Industry analysts say the White House’s intervention could set an important precedent for future AI releases. As AI models become more powerful, phased deployments, enhanced safety evaluations and closer collaboration between technology companies and governments may become standard practice to ensure innovation proceeds without compromising security or public trust.