Editorial illustration for OpenAI's Astra AI model family fuels Sam Altman's 2026 AGI timeline
Altman: OpenAI's Astra Model Targets AGI by 2026
Sam Altman is putting a date on artificial general intelligence, sort of. The OpenAI CEO told TIME's Alex Heath, who spent more than two weeks interviewing over 20 executives, employees, investors and rivals for an extensive report, that OpenAI will have an internal system he'd call AGI by the end of 2026. The catch is his definition, which he admits the company hasn't hit yet.
Chief Research Officer Mark Chen puts the number at 80% of the way there. Co-founder Greg Brockman thinks history will mark this period as the moment AGI actually showed up.
The confidence traces back to Astra, OpenAI's upcoming model family, first
OpenAI CEO Sam Altman is convinced his company will have an internal system he'd call artificial general intelligence (AGI) by the end of the year, even though they're "not quite yet" there.
Why this matters
Altman's 2026 timeline hinges on his own definition of AGI, which is exactly the problem. Pachocki calling Astra an "automated research intern" that already meets an internal bar tells us OpenAI is grading its own homework, with no external benchmark developers or researchers can check. For founders building on OpenAI's stack, that's a signal to treat any AGI announcement next year as a marketing claim first, a technical one second.
Researchers should watch for what Astra can actually verify or reproduce, not what Altman says it can "invent." The gap between "hits our internal bar" and "invents new things that matter" is where real scrutiny needs to land, and TIME's reporting shows even OpenAI's own leadership can't agree on which side of that line Astra sits. Until there's a shared, external definition of AGI that outside labs and independent evaluators sign off on, every date OpenAI floats, 2026 included, is a claim about internal culture and confidence, not a verified capability. Watch for who outside OpenAI gets to test Astra before that date arrives.
Related Reading
Browse all OpenAI news →
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups DeepSeek Seeks More Capital Weeks After USD 7B Funding Round
Related Topic OpenAI researcher quits, citing distrust over ad‑driven engagement metrics
LLMs & Generative AI
Related Topic OpenAI launches GPT-Image 1.5 with precise editing for enterprise visuals
LLMs & Generative AI
From Archives AI agents can propose DNS changes but can't approve them, Tenet finds
From Archives Meta Scrapped AI Layoffs Hours Before They Were Set to Begin
From Archives OpenAI's Data Center Strategy Lead Departs Amid Industry Talent Crunch
From Archives Russian Influence Campaign Used ChatGPT to Push Pro-Kremlin Posts
LLMs & Generative AI
Common Questions Answered
What is Sam Altman's timeline for OpenAI achieving AGI?
Sam Altman has stated that OpenAI will have an internal system he would call artificial general intelligence (AGI) by the end of 2026. However, he acknowledges that the company hasn't fully achieved AGI yet, with Chief Research Officer Mark Chen estimating they are currently 80% of the way there.
How does OpenAI's Astra AI model family relate to the AGI timeline?
The Astra AI model family is central to OpenAI's 2026 AGI timeline, with the company describing it as an 'automated research intern' that already meets their internal AGI bar. This positioning suggests Astra represents a significant step toward their definition of AGI capabilities.
Why is Sam Altman's AGI definition considered problematic?
Altman's AGI definition is problematic because OpenAI is essentially grading its own homework without external benchmarks or independent verification from other researchers and developers. This lack of external validation means any AGI announcement should be treated as a marketing claim first and a technical achievement second, according to the article.
What does Mark Chen's 80% assessment reveal about OpenAI's AGI progress?
Mark Chen's statement that OpenAI is 80% of the way to AGI suggests the company has made substantial progress toward its internal definition of AGI, though significant work remains. This assessment indicates that OpenAI believes it is relatively close to achieving its stated AGI goals by the end of 2026.
What should developers building on OpenAI's stack consider regarding the 2026 AGI announcement?
Developers should treat any AGI announcement from OpenAI in 2026 as a marketing claim first and a technical one second, given that the company is using its own internal definition without external verification. This cautious approach is recommended because OpenAI lacks independent benchmarks or external researcher validation for its AGI claims.
Further Reading
- OpenAI announces its "next major model" Astra by dropping ten previously unsolved math solutions - The Decoder
- OpenAI's timeline is made public: from 2026 to 2028, the strategy ... - 36Kr
- Sam Altman admits he's been 'very wrong' on the timeline that he always says is the goal of OpenAI - Times of India
- Sam Altman's 2026 Strategy: AGI Bets, $500B Stargate & OpenAI ... - Luminix
- OpenAI roadmap revealed: AI research interns by 2026, full-blown AGI researchers by 2028 - TechRadar
