Founders Anna Goldie and Azalia Mirhoseini of Ricursive Intelligence assert that advancements in AI chip design could play a pivotal role in addressing the ongoing compute crisis. During their presentation at TechCrunch Disrupt on Wednesday, they elaborated on how their startup is pioneering AI technology capable of designing chips, learning from each iteration, and enhancing future designs, all aimed at automating and expediting the chip design process.
“There is this understanding that in the next five years we are going to have at least 10,000x more demand for compute, and that’s somewhat potentially conservative,” Goldie remarked during her speech. “So we don’t want to build 10,000 times more data centers. We think that we need to help by making chips that are more efficient, that use the bandwidth in those data centers more efficiently, that use the wafers and fabs more efficiently, because even if we make [data centers] in space or something, I don’t know, you won’t see the sun anymore.”
While the prospect of AI autonomously designing chips may evoke apprehension in some, Goldie contends that allowing AI to manage chip design will enable humans to concentrate on developing applications and exploring new avenues that can enhance their computational capabilities. “I think the way everything is shifting is that humans bring their great ideas and creative thoughts, and execution is handled by AI,” Goldie stated. “And this is happening in chip design, software development, and other areas. And that’s something that recursive does for chip design.”
The founders assert that AI is revolutionizing conventional chip design methodologies by uncovering unconventional layouts that can surpass traditional designs. For instance, AlphaChip, the AI system previously developed by the founders, generated a circular, “donut-shaped” configuration of chip components that initially astonished engineers but ultimately demonstrated its effectiveness. They clarified that their focus is on optimizing entire chips rather than adhering to the conventional method of dividing them into numerous smaller blocks. This strategy enables AI to produce complex patterns, including wave-like designs, which the duo indicated they intend to elaborate on in the future.
Furthermore, the founders posited that chips designed by AI will facilitate the creation of a diverse range of chips that may not have been feasible previously. “If it takes six months or a year or two years to design a new chip, then you really have to be extremely selective about the chips that you design,” Goldie explained. “But if you can design a chip in like a week, and then say for manufacturing, 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, and we can be more energy efficient.”
