With the rise of AI, energy and thermal efficiency have once again become pressing issues for companies that use and build chips. The skyrocketing demand for hardware to run AI models drives up energy bills, as these servers require massive chip counts and huge cooling setups.
Vaire Computing, based in London and Seattle, is betting that reversible computing will be the way forward. It has now raised $4 million in a seed round to work on building silicon chips that will consume negligible amounts of power and generate little, if any, heat. The round was led by deep tech fund 7percent Ventures and Jude Gomila, the co-founder of Heyzap. The company previously raised $500,000, so this round brings its total funding to $4.5 million.
In reversible computation, instead of performing a computation in only one direction (inputs followed by outputs) and then feeding the output to a new computation and starting it again, the computation can be performed in both directions (known as “time-reversible ” calculation ). Energy is effectively trapped inside the chip instead of being released as heat. The theory is that this method will generate negligible amounts of heat, greatly reducing energy consumption. (A better explanation of its potential is found in this essay by Azim Azhar and David Galbraith.)
Vaire Computing was founded by serial entrepreneur Rodolfo Rosini and Hannah Earley, a researcher at the University of Cambridge who works on “unconventional computing” such as reversible and molecular computing.
During a conversation, Rosini told me, “Nearly 100% of the energy in the chip is ultimately dissipated as heat. So you basically lose it. But in a reversible chip, you never actually dissipate that energy. You don’t let the energy turn into heat and recycle it internally. This means two things happen: First, the chip doesn’t heat up, and second, you only need a small amount of power to make it work. So it uses almost no energy other than the same amount of energy it just recycled.”
The concept of reversible computing is not new, and there are many challenges before Vaire’s chips become a reality, but Rossini believes that the transition to this new approach to computing will not be too different from how we moved from incandescent light bulbs to LEDs. “The similarity is between an old incandescent light bulb and LEDs,” he said. “LEDs are cooler and more efficient, and there’s a cluster of them… It’s almost identical to the reversible calculation. You don’t have a single core that’s super fast, you have many smaller cores that are each super efficient.”
He says a big advantage of chips that can do reversible calculations would be their ability to be used in general applications, just as normal processors are used today. “Other types of chips are domain specific, but with computing you can do anything… We can also build a CPU or a GPU and it will look like any other chip.”
Asked why funding in space is so low if the technology is as revolutionary as it sounds, Rosini said: “Because the amount of money that went into reversible computing and alternative chip architectures is next to nothing,” he said, pointing to the billions spent on quantum computing, photonics and GPUs.
“If you go outside of these well-trodden areas and talk about building brand new architecture, there is absolutely no one to fund it. Second, we don’t really need a lot of money to make the first chip and prove the technology… Once we prove that, we’ll need a much bigger round to build a chip,” he added.
For his part, Earley believes that reversible computation can be used to make the most powerful computers. “I got involved in this field during my PhD in 2016,” she said. “By chance, my PhD supervisor sent me the dissertation to a friend of mine who was in a group at the University of Florida that was doing reversible calculus. It got me interested in how I could apply it to my research field at the time, which was molecular programming. I’ve come to think that reversible computing is interesting in its own right, especially as it can make the most powerful form of computing possible. After I finished my PhD, I met Rudolfo and we realized that we had the same vision.”
“Vaire Computing is different because its technology is innovative at a fundamental level, positioning the company extremely well to capture a huge chunk of the future AI chip and ultimately computer processor market,” said Andrew J. Scott, founding partner at 7percent Ventures in a statement.
Also participating in the round were Seedcamp, Clim8, Tom Knight (inventor of modern reversible computing), and Jared Kopf, founder of Ramble.ai.
In addition, Vaire hired Mike Frank, a renowned researcher in reversible computing, as the company’s senior scientist.
Vaire recently became one of only 10 companies named in the second UK cohort of Intel Ignite, Intel’s global startup accelerator program for early-stage deep technology start-ups.
