OpenAI says its new model has achieved the AI industry’s most sought-after milestone.
“Welcome to the AGI era,” company president Greg Brockman said at the end of a press conference announcing the arrival of OpenAI’s Astra model on Thursday.
AGI stands for artificial general intelligencea vague concept that has nonetheless become a core goal for AI development. OpenAI defines it as “highly autonomous systems that outperform humans at the most economically valuable work.” The company presented the Astra as a major breakthrough in research and a shift in what is possible to delegate to AI. It will be rolled out to customers this week.
Brockman has described achieving AGI as a gradual process. On Thursday’s call, he said he thinks people in the future will look back and think AGI was founded “around this time,” and I think it might be about this model.
“For me personally, I think we’re there,” he said.
GPT-6 Astra comes about a year after OpenAI launched its GPT-5 model family and about two months after its most recent upgrades, called GPT-5.6. The company called the Astra “the most intelligent and tuned model in the world” and a “step change in professional work.”
The model will also improve ChatGPT’s “computing” capabilities, the company said. OpenAI employees told Business Insider last month that these tools – which perform tasks autonomously on a user’s desktop computer or Internet browser – have reached a tipping point. The blog post stated that Astra can fill out forms, create documents, schedule appointments and operate software.
While successful “computing” AI agents have long been a goal of OpenAI and other tech giants, the technology is tricky because there is such a wide variety of people’s desktop work, and so far the tools have been quite slow.
“Anything you can do on a computer, Astra can do for you. Fast,” OpenAI wrote on X.
OpenAI previously said it was delaying the release of Astra until September after improvements were identified cybersecurity capabilities.
The company wrote earlier this week that it needed to ensure that Astra’s protections block its use by bad actors for hacking and that the model itself is tuned to do what’s asked of it.
Like Brockman, other technology leaders have prioritized AGI as a goal. Co-founder of Google DeepMind Demis Hassabis said at the company’s conference in May that he expected AGI to arrive as early as 2030, adding that humanity is “at the foot of the singularity.”
Sam Altman, CEO of OpenAI, recently told the Sources podcast that AGI is at best “a very ill-defined term. I was going to say it’s an irrelevant marketing term.”
Do you have a tip? Contact this reporter by email at scouncil@businessinsider.comor via text, Signal, Telegram or WhatsApp at 415-757-8198. Use a personal email address, a non-real Wi-Fi network, and a non-work device; here is our guide to share information securely.
