OpenAI has launched GPT-6 Astra, and Greg Brockman announces that AGI has come.
SAN FRANCISCO, CALIFORNIA – OCTOBER 03: Greg Brockman, Co-founder, Chairman, and CTO of OpenAI, presented on stage during TechCrunch Disrupt San Francisco 2019 at the Moscone Convention Center on October 03, 2019, in San Francisco, California. (Photo by Steve Jennings/Getty Images for TechCrunch)
OpenAI's president has announced that artificial general intelligence (AGI) has arrived. According to the Washington Post, Greg Brockman indicated that the company's new Astra model meets this criterion. OpenAI characterizes AGI as systems that exhibit greater intelligence than humans. "I leave it to the reader to determine if this qualifies in their view," Brockman stated. "I believe we have reached that point."
The model was made available to business users in the Daybreak program on Thursday. Bloomberg reports that a version with enhanced cybersecurity measures is scheduled for paid ChatGPT subscribers. Both versions restrict access to advanced cyber capabilities. OpenAI has asserted that Astra has attained its critical cybersecurity threshold, and Sam Altman mentioned that its release has taken longer than anticipated.
This classification is related to the model OpenAI delayed in August, which achieves this status when it can discover and develop effective zero-day exploits against fortified systems autonomously. Astra has been designed to utilize computers similarly to humans, capable of managing financial modeling, tax preparation, and video game development with minimal human intervention.
Prior to its release, the US government evaluated it. Brockman noted that there were no recommendations for changes from their review, which he presented as a reassurance. However, this claim might be weaker than it appears.
The executive order from June permits developers to provide agencies access to models for up to 30 days before a widespread launch and clearly states that it does not authorize mandatory licensing, preclearance, or permitting. Participation is voluntary, and the NSA director determines which models qualify as covered frontier models via a classified benchmarking process. No approvals were withheld because none were offered.
Europe also does not pre-clear models. The Commission obtained the authority to carry out evaluations on August 2, driven by inadequate documentation or alerts from its scientific panel. It can require access through APIs or source code. These powers are reactive rather than preventative. Additionally, starting from September 11, manufacturers of products with digital components must report actively exploited vulnerabilities within 24 hours of becoming aware.
Thus, no approval for this model was granted in either jurisdiction. OpenAI has been revising its framework following the Hugging Face breach, and its own classification remains the authoritative one.
Get the TNW newsletter
Receive the key technology news in your inbox weekly.
Other articles
T-Mobile's Gopalan focuses on home internet and AI following a 25% decline.
Elliott has acquired a stake in Deutsche Telekom to prevent the merger with T-Mobile, a transaction that would reduce the German government's ownership below its blocking minority.
.jpg)
The top 10 autonomous pentesting tools, ranked by exploit verification (2026)
Ten autonomous pentesting platforms were evaluated based on proof-of-exploit, attack-chain depth, and coverage. Astra Security stands out for web and API exploitation, utilizing a separate validator that re-exploits each finding before it enters your queue.
New Jersey has recently authorized affordable plug-in solar panels for balconies.
New Jersey has prohibited landlords and municipalities from preventing plug-in solar installations. In 2024, Germany granted this right to tenants, and just last week, Britain legalized such systems.
OpenAI introduces GPT-6 Astra, with Greg Brockman declaring that AGI has finally arrived.
The president of OpenAI states that Astra meets the criteria for AGI. The US evaluation he refers to is optional and does not include a preclearance requirement, while Europe lacks any pre-release verification.
Why video serves as the initial frontier as AI begins to comprehend the physical world
With 562 million surveillance cameras currently in place globally and two-thirds of them now equipped with deep-learning analytics, Lumana is wagering that transforming existing video feeds into searchable and intelligent sources is the quickest route for physical AI to achieve scalability.
Sam Altman states that AI developers have struggled to effectively communicate the advantages of the technology.
Sam Altman notes that the AI industry has poorly communicated its message. In Europe, the limitation on data centers is determined by grid capacity and reported energy data, rather than public perception.
OpenAI has launched GPT-6 Astra, and Greg Brockman announces that AGI has come.
The president of OpenAI states that Astra meets the criteria for AGI. The US assessment he refers to is voluntary and does not require preclearance, while Europe lacks a pre-release evaluation.