Cortical Labs CEO Dr. Hon Weng Chong said biological neurons are 5,000 times more stable and sample efficient than traditional GPU-based AI systems, reshaping the future of reinforcement learning.
Chong warned against creating conscious systems, stating, “You do not want to create conscious systems because ethically a conscious system has the ability to suffer.” The company’s CL1 platform allows developers to program biological neurons using Python, bypassing custom hardware.
The firm is building the world’s first biological data center, which requires no special cooling and operates without affecting the energy budget. Each CL1 supports up to 2 million neurons, with 200,000 considered commercially viable for learning and training.
Chong emphasized that even simple organisms like flies possess generalized intelligence-something current machines lack-pointing to biology as a model for next-generation AI.