A team of scientists, led by Johns Hopkins University, has proposed the development of a biological computer that could surpass silicon-based machines in performance and energy efficiency.
The computer will be powered by millions of human brain cells, arranged in arrays of brain organoids, which are small three-dimensional neural structures grown from human stem cells. The organoids will be connected to sensors and output devices, and trained using techniques such as machine learning and big data.
The researchers have published a detailed roadmap in the journal Frontiers in Science, outlining their vision for what they call “organoid intelligence”. This ultra-efficient system aims to solve problems that are beyond the capabilities of conventional digital computers, while also supporting the development of neuroscience and medical research.
Although similar to quantum computing in ambition, the project raises ethical concerns regarding the “consciousness” of brain organoid assemblies.