The core of this system is the Biohybrid Processing Unit (BPU), a rack-mountable device that fits into standard data center infrastructure. Inside, a living neural organoid is sustained by specialized hardware that provides life support and signal processing. To make the tissue responsive, Intactis mapped over 150 neural relationships, allowing the system to interpret electrical inputs as game moves. CEO Daniel Rodriguez-Granrose asserts that biocomputation is no longer a black box, noting that they have successfully cataloged the tissue's responses to dictate precise moves within the game.
This technology addresses the escalating energy crisis fueled by modern AI development. While a traditional supercomputer may consume 20 megawatts, the human brain operates on approximately 20 watts. Intactis projects that their BPU could reduce energy costs by 95% and lower the physical data center footprint by 88% compared to exaflop-scale silicon. The company is currently offering Cloud Biocompute as a Service to developers in robotics and artificial intelligence, supported by over $1 million in initial funding and a leadership team with a history of significant industry exits.

Comments (0)
No comments yet. Be the first!