According to Goldman Sachs, the GPUs needed to run AI models consume enormous amounts of energy, which will increase electricity demand by 160% by 2030.Vishal Sarin, an analog circuit and memory designer, founded Sagence AI to create more efficient alternatives to this problem, as reported by TechCrunch.
The startup develops analog chips that, unlike digital chips, can represent data over a range of values and do not require data transfer between memory and processors, which reduces power consumption and improves speed.
Although analog chips were popular from 1935 to 1980, their energy efficiency is being reconsidered due to the limitations of digital chips, such as the need for multiple components and memory constraints.
more
Chip wars: Japan plans to turn the tide Shutterstock/Riopatuka The difference between Sagence chips Sagence chips, called "in-memory chips," can perform calculations with fewer modules and greater data density, providing superior performance and lower latency for AI applications. Despite their advantages, analog chips are more difficult to manufacture and program with high precision. Sarin believes Sagence chips will complement digital chips and accelerate specific applications in servers and mobile devices. The company, which plans to launch its chips in 2025, has already invested US$58 million and is preparing to expand its team. With the rise of semiconductor startups and the growing demand for AI solutions, Sagence can profit as long as it can prove the superiority of its chips in terms of power consumption and efficiency.
Sagence chips will be launched in 2025 - Image: 3dartists / Shutterstock.com