Do you understand quantum parallel repetition? What about quantum complexity, lattice cryptography, or extremal combinatorics?
The new OpenAI model Astra knows all about them.
Recently, OpenAI confirmed the existence of Astra, calling it “our next major model.” On Aug. 1, the ChatGPT-maker revealed that “an internal version of Astra” solved 10 major open math problems, some of which have been unresolved for decades. Then, on Aug. 7, OpenAI announced that Astra had developed such advanced cyber capabilities that new security controls were necessary and that some internal development work would be paused.
“These results, in addition to expert assessments, have led us to conclude last night that we cannot rule out critical cyber capabilities under our Preparedness Framework.”
OpenAI’s Preparedness Framework is the company’s “approach to tracking and preparing for frontier capabilities” that could be dangerous. It tracks risk levels in three categories: biological and chemical, cybersecurity, and AI self-improvement.
Previously, GPT-5.6-Sol was evaluated as a “high” risk in cybersecurity, and so Astra could be the first OpenAI model to be classified as a “critical” risk.
“Under our Preparedness Framework, a model reaches the Critical cybersecurity threshold if it can identify and develop functional zero-day exploits of all severity levels in many hardened real-world critical systems without human intervention, or can devise and execute end-to-end novel strategies for cyberattacks against hardened targets given only a high level desired goal,” OpenAI’s blog post states.
OpenAI’s testing now shows enough progress on agentic coding and developing zero-day exploits (previously undiscovered vulnerabilities in software or hardware) that the company is stepping back to strengthen security around Astra. To do this, the company is tightening sandboxes that keep the model contained and monitoring its Chain of Thought to “interrupt high risk activity.”
OpenAI also said that it’s committed to working with “relevant government agencies and select AI safety organizations to test the capabilities for this model.”
Frontier AI models have progressed rapidly in the cybersecurity domain this year. Some zero-day bug bounty. programs have been forced to shut down due to the deluge of AI-discovered bugs, as Mashable has reported.
Until recently, the prospect of swarms of AI agents conducting relatively autonomous cyber attacks on critical infrastructure was mostly hypothetical. That risk seems much more real thanks to unreleased models like Claude Mythos and OpenAI Astra.
Our big Guessing Game is back! Enter now for a chance to win an Apple Watch.
What else do we know about OpenAI Astra?
Precious little, to be honest. Prior to the most recent announcement, OpenAI had released few concrete details, besides its internal name, Astra, and the fact that it was in testing to become OpenAI’s next major model.
Bleeping Computer recently described Astra as “a powerful model that allows AI agents to collaborate on different parts of a larger problem.” That suggests it’s designed for agentic work and can “tackle complex, long-running tasks,” also per Bleeping Computer.
At this point, it’s unclear if Astra will eventually be released as GPT-5.7, the beginning of GPT-6, or something else entirely. If it truly has critical cybersecurity capabilities, it may never be fully released to the public, as with Anthropic’s Claude Mythos.
However, let’s look again at how OpenAI initially described Astra: “The results were achieved by an internal version of Astra, our next major model.”(Emphasis added by Mashable.) That certainly suggested that OpenAI was preparing Astra for release sooner rather than later. However, in its Aug. 7 blog post, OpenAI’s language had shifted, describing Astra as “one of our upcoming models.” (Emphasis added by Mashable.)
The company has also so far declined to answer our questions about Astra, but we’ll update this story if we receive more information.
Mashable Light Speed
However, we can make some additional deductions. Based on OpenAI’s original blog post about Astra, “Ten advances in mathematics and theoretical computer science,” it’s safe to say that the new Astra model has made significant advances in science and mathematics, as well as in cybersecurity.
So, we talked to a mathematician about what Astra means for the future of AI and science — and what it doesn’t mean.
New Anthropic and OpenAI models are really good at coding and math
Credit: Photographed by Joseph Maldonado / Mashable Composite by Rene Ramos
Earlier this year, Anthropic announced that its unreleased Mythos model was so good at hacking that it was too dangerous to release to the public. At the time, we questioned this narrative, as the company’s warnings effectively doubled as PR for itself. However, we can’t deny that the latest frontier models from Anthropic and OpenAI have gotten remarkably good at cybersecurity, agentic coding, and discovering zero-day bugs.
More recently, OpenAI and Anthropic have shown progress on research-level mathematics as well.
First, OpenAI released a disproof of the Erdős unit distance conjecture in May. Then, an Anthropic-linked mathematician casually announced that he used Fable 5 to disprove the Jacobian Conjecture, one of the infamous Smale’s problems in mathematics. Now, OpenAI has announced that Astra solved 10 more open problems in mathematics.
It’s a potentially paradigm-shifting moment, though this is hardly proof that artificial general intelligence or the singularity is nigh.
First, most of these math accomplishments take the form of disproofs and counterexamples, which aren’t as impressive as positive proofs that greatly expand our understanding of the universe, like, say, Andrew Wiles’ proof of Fermat’s Theorem.
When Fable 5 disproved the Jacobian Conjecture, I spoke to Columbia professor Andrew Blumberg, who is also on the board of directors of the First Proof project, which tested the capabilities of frontier large language models in solving research-level mathematics.
At the time, he told me that Fable 5’s feat “did not cause me to update my priors about what AI can and can’t do.” Blumberg added, “This is exactly the kind of thing I would expect AI to be able to do. If there was a counterexample that was concise and easy to state that people haven’t found because it’s a pain to search through all this stuff, AI will find it.”
Ultimately, Blumberg said that Fable 5’s counterexample to the Jacobian Conjecture didn’t necessarily teach us anything new or exciting about the world, even though it was very impressive.
I went back to Blumberg to ask about OpenAI and Astra’s latest work, and he said his priors still haven’t changed — AI is a very useful tool for advanced science, but Astra’s math results don’t suggest AI is ready to replace human scientists and mathematicians.
“If you look at what these are, they’re pretty short. They are either counterexamples or they are small, extremely clever constructions that build on known things, and it’s super cool, right? I want to be clear: If a person had done many of these things, that person would be justly lauded for their achievement, and so [this work is] great, but they don’t change my priors.”
And, as many people have pointed out (including, most recently, data scientist Nate Silver), we have to put these results in perspective. While OpenAI was keen to point out that these problems were solved with just $2,000 worth of tokens, that doesn’t account for the trillions spent on AI technology in recent years.
“When you hear about the amount of money that’s being poured into these machines, and you think about what it would be like if we spent a trillion dollars on hiring and training mathematicians and paying the best mathematicians football players’ salaries to do nothing but solve these problems, I think we’d see a shitload of progress,” Blumberg told Mashable.
This Tweet is currently unavailable. It might be loading or has been removed.
At the same time, the prospect of Astra-level models being in the hands of every scientist, physicist, and mathematician on Earth is truly exciting. Likewise, the prospect of Astra-like tools being available to hackers and other bad actors is truly concerning.
When will GPT Astra be released?
Previously, we speculated that GPT-Astra would be released by the end of the year, but now that OpenAI is delaying development for security reasons, it’s much harder to say. As we mentioned earlier, it may never be fully released.
There are other factors that could delay its release as well.
First, the White House is reportedly close to finalizing a voluntary AI framework for testing new frontier models before they’re released publicly. OpenAI has now publicly committed to working with relevant government agencies before releasing Astra. And if Astra is deemed to have “critical” capabilities, then additional time would be required to develop a safe-for-public-consumption version of the model.
There is reason to believe that Astra is far along in its development cycle, however. The Information recently reported that OpenAI is actively previewing Astra in Washington, D.C., and if Astra is already generating major results in math, it’s likely fairly far along in testing.
Major AI companies like OpenAI and Anthropic have been releasing major updates to their models every few months, with entirely new model families coming out every one to two years. GPT-5 was introduced in August of 2025, which means we’re due for GPT-6 any time now.
So far, OpenAI hasn’t said anything about a release timeline.
Its most recent blog post about Astra concludes: “We’re committed to working alongside governments, safety institutes, and civil society to ensure that the frontier capabilities of models like Astra, and those that follow, are deployed responsibly and broadly for the benefit of all humanity.”
Disclosure: Ziff Davis, Mashable’s parent company, in April 2025 filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.
