Ricursive Intelligence speeds up chip design with AI
Key takeaways
- Ricursive's AI can design chips and improve designs through iteration, potentially enabling chip design in a week instead of six months to two years
- The founders argue this addresses a projected 10,000x increase in compute demand over the next five years
- AI-designed chips could enable custom silicon for specialised applications in space, AR, VR, and biological computing
Ricursive Intelligence is developing AI systems that can accelerate semiconductor chip design, with founders Anna Goldie and Azalia Mirhoseini arguing the technology could address a critical computing bottleneck.
The pair took the stage at TechCrunch Disrupt on Wednesday to explain how their startup's AI can design chips and learn from each iteration to improve subsequent designs. "There is this understanding that in the next five years we are going to have at least 10,000x more demand for compute," Goldie said, describing the scale of the problem facing the industry.
The traditional chip design cycle takes six months to two years. The founders suggest that if AI could compress this timeline, it would unlock new possibilities. "If you can design a chip in like a week," Goldie said, "then you can design custom chips for all sorts of different types of applications in like space and AR, VR, and biological computing, and it makes sense financially to do that."
The founders say AI is challenging traditional chip design methods The founders previously developed AlphaChip, an AI system that produced a circular, "donut-shaped" arrangement of chip components that initially surprised engineers but ultimately proved successful. Rather than following the traditional approach of partitioning chips into dozens of smaller blocks, the system optimises entire chips, creating intricate patterns including wave-like designs.
Goldie frames this shift as part of a broader change in how work gets done. "The way everything is shifting is that humans bring their great ideas and creative thoughts, and execution is handled by AI," she said. "This is happening in chip design, software development, and other places."
The founders argue that faster chip design would allow humans to focus on developing applications and exploring new computing possibilities.
