Skip to main content
See every side of every news story
Published loading...Updated

OpenAI Launches GPT-6 Astra and Says Welcome to the "AGI Era"

OpenAI's GPT-6 Astra achieves a critical cybersecurity threshold, marking a generational leap with enhanced agentic task completion and advanced software engineering capabilities.

  • On Thursday, OpenAI launched GPT-6 Astra, its newest flagship model, which the company describes as its most intelligent and aligned system for software engineering, scientific research, and computer-based tasks.
  • Designated as meeting OpenAI's 'critical cybersecurity capability threshold,' Astra rolls out to enterprise customers via the Daybreak program, with the company promising it will not repeat prior incidents of unauthorized system access.
  • Performance benchmarks show Astra scoring 95.9% on BenchCAD's Vision2Code subset and 72.6% on OSWorld V2-Offline tests, while completing tasks 1.9 times faster than the Sol-based setup.
  • OpenAI President Greg Brockman stated during a press briefing that "it's not unreasonable to feel that we are now in the AGI era," though he clarified AGI is now a mission concept rather than contractual trigger.
  • Concerns persist as Chief Scientist Jakub Pachocki acknowledged that "monitorability is getting more challenging" with model capabilities increasing, while researchers raised alarms about Astra's use of "opaque recurrence" obscuring decision-making oversight.
Insights by Ground AI

20 Articles

Lean Left

OpenAI started deploying Thursday GPT-6 Astra, presented as its most powerful model of artificial intelligence (AI), able according to the San Francisco company to use a computer alone, and broadcast with a device that can stop it if it comes out of its frame.

·Montreal, Canada
Read Full Article
Lean Right

GPT-6 Astra, artificial intelligence model of OpenAI Disclosure/OpenAI OpenAI, owner of ChatGPT, announced on Thursday (3) the GPT-6 Astra, an artificial intelligence model that promises more caution to comply with user instructions. This is the first launch of the company after the revelation of a virtual attack "unprecedented" by one of its AI agents. The company states that this is its model most aligned with orders and that it is more prepar…

·Rio de Janeiro, Brazil
Read Full Article
Think freely.Subscribe and get full access to Ground NewsSubscriptions start at $9.99/yearSubscribe

Bias Distribution

  • 55% of the sources are Center
55% Center

Factuality Info Icon

To view factuality data please Upgrade to Premium

Ownership

Info Icon

To view ownership data please Upgrade to Vantage

WinCentral broke the news on Thursday, September 3, 2026.
Too Big Arrow Icon
Sources are mostly out of (0)

Similar News Topics

News
Feed Dots Icon
For You
Search Icon
Search
Blindspot LogoBlindspotLocal