FrontierThe story, in brief

OpenAI says 80 to 90 percent of its research already targets GPT 7 and beyond

80 to 90 percent. That's how much of OpenAI's research budget is already locked on GPT 7, 8, and beyond—not on making GPT 6 faster or cheaper.

Illustration of a transparent lens revealing connected networks across layers of paper.
Exploring the next frontier of AI research.AI illustration by KeyNews
The KeyNews take

Why it matters

OpenAI has publicly committed its near-term research to models two generations ahead, signaling where frontier capability investment is flowing and implicitly deprioritizing incremental improvements to current-generation models. This shapes what practitioners should expect from near-term product updates and where the lab race is actually being run.

The key facts

5 to know
  1. 80-90% of OpenAI research targets GPT 7, GPT 8, and beyond

  2. Single-generation improvements intentionally treated as short-term bets

  3. Boris Power (Head of Applied Research) made the statement

  4. Power stated most users don't know what they can do with AI, positioning capability discovery—not raw model performance—as the bottleneck

  5. Published September 27, 2026

Go to the source

The Decoderthe-decoder.com

Publisher excerpt: Boris Power, OpenAI's Head of Applied Research, says 80 to 90 percent of the company's research goes toward GPT 7, GPT 8, and beyond. Improvements within a single generation are intentionally short-term bets. But Power doesn't see model performance as the biggest problem. Most users don't even know…
Read original report
Back to today's editionMore frontier news

Keep reading

Related stories

More from Frontier