Topics: AI | Hardware | Science & Research | Startups | Technology
Jeff Shainline, co-founder and CEO of Great Sky, argues that superconducting, brain-like hardware could outperform GPU-era AI for fast inference.
●
SOENs split computation, memory, and communication across superconducting circuits and optical links to remove the von Neumann bottleneck.
●
Great Sky says its chips run about 2 million times faster than biology and have shown roughly 100x speedups on small language models.
●
The near-term targets are video, speech, fusion control, and other low-latency workloads where rapid inference has direct value.
We build neurons and synapses like the ones in the brain, but everything in our hardware happens about 2 million times faster than biology — Jeff Shainline