OpenAI has agreed to delay the release of its new artificial intelligence model, GPT-5.6, following a request from the Trump administration due to concerns over potential security issues. According to The Information, during an internal company Q&A session, OpenAI CEO Sam Altman announced that GPT-5.6 would only be made available in a limited preview format, where its accessibility would be restricted to a select group of enterprise customers. Moreover, during this preview phase, each customer’s access to the technology will be approved by the Trump administration on a case-by-case basis.
This cautious rollout of GPT-5.6 by OpenAI stands in contrast to the treatment of their AI competitor, Anthropic, which faced a more stringent approach from the Trump administration. Earlier in the month, Anthropic was subjected to an ultimatum that forced it to cut off access to its AI models, Mythos 5 and Fable 5, to all users except U.S. citizens. This came as part of an export control directive intended to prevent foreign nationals from accessing these advanced technologies. Moreover, the directive impacted even Anthropic’s employees who were non-U.S. citizens, prohibiting them from accessing the company’s own AI tools.
This disparity in regulatory measures imposed on OpenAI and Anthropic by the Trump administration indicates a potentially uneven application of AI regulations, which has raised concerns across the technology industry. The worry stems from an apparent deviation from earlier promises made by the Trump administration, which professed a “speed wins” approach aimed at bolstering the American AI exports program and encouraging rapid development and dissemination of AI technologies.
The decision to limit access to GPT-5.6 specifically to enterprise customers and to require government approval on a case-by-case basis showcases a more controlled and vigilant approach to how potent AI tools are managed and distributed in the U.S. This strategy reflects growing governmental awareness and caution regarding the implications of increasingly powerful and autonomous AI systems, particularly in the realm of national security.
The varied treatment between OpenAI and Anthropic also highlights broader trends in governmental control over AI, suggesting an evolving regulatory landscape that may require AI firms to navigate an increasingly complex array of requirements and constraints. The consequence is a technology sector that must remain agile and responsive to both rapid advancements in AI and shifting policies that attempt to govern its application and reach.
Overall, the Trump administration’s request for OpenAI to delay the full release of GPT-5.6, allowing only limited access under strict supervision, serves as a significant example of the government’s growing involvement and possibly cautious stance towards the progression and export of AI technologies. This move, paralleled by stringent controls on other AI entities, underscores the importance and challenges of balancing innovation with security in the rapidly evolving domain of artificial intelligence.
Read the full post on theverge.com
wnds6s