Amazon reportedly scales back its Nova AI models and bets on a new Frontier research team
Quick Answer
Amazon is scaling back its Nova AI models, including Nova Premier and Omni, while shifting focus to the Frontier Model Research team under Pieter Abbeel.
Quick Take
Existing models will remain operational but not actively developed, following layoffs in the AGI division and the closure of the AGI Lab.
Key Points
- Nova models, including Premier and Omni, will be maintained but not developed further.
- The shift follows layoffs in the AGI division and closure of the AGI Lab.
- Resources are redirected to the Frontier Model Research group for new developments.
- A new foundation model is expected to be announced at the re:Invent conference.
- Amazon remains a major investor in Anthropic and OpenAI, with investments potentially exceeding $100 billion.
📖 Reader Mode
~1 min read
Jul 28, 2026
Amazon is overhauling its AI strategy and scaling back most of its in-house Nova models, Business Insider reports, citing insiders. The cuts hit the flagship Nova Premier and Omni models, the Reel video model, and the Canvas image generator. The models stay online in "keep the lights on" mode for existing customers but are no longer actively developed. The pullback follows layoffs in the AGI division and the closure of the AGI Lab Amazon founded in 2024 after acquiring Adept. Amazon first unveiled the Nova models in December 2024.
Amazon isn't exiting the AI race, though. Resources now flow to the Frontier Model Research group under Pieter Abbeel, who joined through the acquisition of robotics startup Covariant. A new foundation model is due at the re:Invent conference this fall. A spokesperson told BI that AI models remain one of Amazon's most important projects. The company is also one of the biggest backers of Anthropic and OpenAI, where investments could top $100 billion.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
