FrontierSeptember 3, 2026via The Decoder
GPT-6 Astra is the first model making OpenAI willing to declare the "AGI era"
Why it matters
GPT-6 Astra represents a genuine frontier capability jump: benchmark dominance across math, coding, and security, plus autonomous vulnerability discovery that crosses into territory OpenAI's safety framework now rates as 'critical.' This is the lab-race inflection point practitioners and enthusiasts have been tracking.
Key signals
- OpenAI releases GPT-6 Astra
- President Greg Brockman declares start of 'AGI era'
- Tops benchmarks in math, coding, cybersecurity
- First model rated 'critical' under OpenAI's safety framework
- Independently discovered two previously unknown zero-day vulnerabilities during testing
- Published: September 3, 2026
The hook
OpenAI just declared the 'AGI era' has begun. GPT-6 Astra found zero-days on its own.
OpenAI has released GPT-6 Astra, its most capable model yet. President Greg Brockman says it marks the start of the "AGI era." Astra tops benchmarks in math, coding, and cybersecurity and is the first model OpenAI rates as "critical" under its safety framework. During testing, it independently found…