OpenAI has opened GPT-5.6 Sol, Terra and Luna to all users after a government-requested limited preview and introduced GPT-Live voice models capable of real-time conversation. The timing follows Anthropic's restoration of its own latest models after a dispute over export controls.

The models reached ordinary users on Thursday, about two weeks after the firm had confined early access to only a "small group of trusted partners" at the request of the U.S. government.

OpenAI added that it does not think this variety of government review "should become the long-term default" since it withholds advanced capabilities from users, developers, enterprises, cyber defenders and international partners.
Tap a lens to see what this story means for you.
Liked this? The Brief brings you the whole day in tech, verified, every morning.
Two minutes, free forever. What's in The Brief →
See what’s happening right now
The Feed runs all day — short, verified briefs the moment they break.
Open the FeedFollow @thecircuitry_
Every story we publish, as it happens. No noise between.
Reader-supported
The Circuitry is a passion project I've always wanted to build, and I love the work behind it.
Running it costs real money. APIs, hosting, time. To keep improving the site and growing this into something useful for everyone, those costs have to be covered.
Any contribution is appreciated. If not, no pressure. Thanks for reading.
OpenAI's chief scientist has proposed shared limits on model perplexity and published internal scores from GPT-3.5 onward. Those scores put GPT-3.5 near 0.99 and GPT-4 near 0.91, and the proposal asks labs not to cross a shared bar until monitoring improves.
OpenAI has cut the average success rate of attacks that extract protected reasoning from an earlier request to 0.2% on GPT-6.2 Astra, from 54.0% on GPT-6 Astra. Current-request attacks are at 0.9%, and selected safeguard improvements are being applied back to GPT-6 Astra.
OpenAI has introduced MedChemBench, a 99-problem test of how models read molecules, explain lab results, and choose the next compound. At the xhigh setting, GPT-6.2 Astra scored 84.5%, while compound choice and property forecasts remain much weaker than structure work.
OpenAI says its models found up to 14% PPA improvement on AMD's flagship GPU design, and AMD verified single-digit gains on key blocks of its next flagship CPU. Some of the changes are already in a design AMD sent for manufacturing.
NATO has used OpenAI for up to 300 billion tokens a month on cyber defense, and core institutions can use advanced cybersecurity models through Daybreak. The access runs through NATO's Communications and Information Agency, and a 2026 exercise trained operators beside AI agents.