The Information Machine
Updated todayNew2 sources

OpenAI AI Models Achieve 14% Chip Design Gains with AMD in Five Months

The gist

The results show AI contributing measurable improvements to chip designs that experienced engineers had already heavily refined. OpenAI applying the same approach to its own custom inference chip indicates the technique is being used in production hardware development.

When stories like this move, I write about them in an email each morning.

One free email each morning, around 8 AM US Eastern.

The full picture

OpenAI's AI models identified up to 14% improvement in power, performance, and area on AMD's flagship GPU IP in under five months, building on designs AMD engineers had already spent years optimizing. Greg Brockman said the AI-driven gains were additive to human engineering work, not a replacement for it. OpenAI applied the same approach to Jalapeño, its first custom inference chip, with gains verified on real silicon. OpenAI is also partnering with Synopsys to integrate its model capabilities with Synopsys's EDA design and validation tools for broader access by chip designers.

How it developed
18 November 2026

Greg Brockman described AI compressing a year of chip improvements into five months, additive to human engineering

Sources

Get the daily email

When stories like this move, I write about them in an email each morning.

One free email each morning, around 8 AM US Eastern.

See the latest edition · Privacy