When Computers Grow Brains: Rethinking AI with Living Cells | Hon Weng Chong | TEDxMelbourne

Quick Overview

Dr. Hon Weng Chong proposes rethinking Artificial Intelligence (AI) by utilizing living brain cells grown on computer chips to perform basic computational tasks, demonstrating that these bio-hybrid systems can learn to control a simple video game, Pong, using significantly less energy than traditional silicon-based AI.

Key Points: The speaker, Dr. Hon Weng Chong, discusses using living brain cells grown on computer chips to create bio-hybrid AI systems. These brain-cell systems can learn to control basic tasks, such as playing the video game Pong, by receiving and sending electrical signals. The energy consumption for these biological systems is vastly lower, running on roughly the same energy as a lightbulb, compared to traditional supercomputers. The brain cells used in the experiment were derived from stem cells, which were reprogrammed using a technique developed by Japanese researcher Professor Shinya Yamanaka. The research raises significant ethical questions regarding who owns the content generated by such systems and the implications for future technology. The goal is to build larger, more complex systems that integrate biology and computation to advance AI understanding.

Context: The presentation takes place at a TEDx Melbourne event, featuring Dr. Hon Weng Chong, who explores the intersection of biology and computing by using living brain cells to create rudimentary artificial intelligence. He contrasts this energy-efficient biological approach with the massive resource demands of current silicon-based AI, framing the discussion around the concept of consciousness and the ethical considerations of bio-hybrid technology.

Detailed Analysis

Dr. Hon Weng Chong introduces the idea of creating Artificial Intelligence not solely with silicon, but by integrating living brain cells grown on computer chips. He highlights that every living organism, including humans, dogs, cats, bees, and octopuses, possesses brains that allow them to survive and thrive, communicating via tiny electrical signals. His lab works on creating these bio-hybrid systems by taking adult cells (such as skin, blood, or cheek cells) and converting them back into stem cells using the Yamanaka factors, a technique that earned Professor Shinya Yamanaka a Nobel Prize. These stem cells are then differentiated into brain cells and grown on a computer chip. The energy cost is minuscule; the system runs on about the same energy as a lightbulb, which is far less than traditional supercomputers. The speaker demonstrates that these cell cultures, connected to a computer chip, can learn to play the classic video game Pong within five minutes, controlling the paddle by sending and receiving electrical signals, effectively teaching the cells a task without a body. This leads to ethical questions about ownership of the resulting computation and what consciousness means in this context. Chong urges the community to consider these ethical implications as this technology advances toward building larger, more complex systems.