-: FOLLOW US :- @theinsaneapp
Researchers from multiple universities have published a paper titled "Brain Organoid Computing for Artificial Intelligence."
-: FOLLOW US :- @theinsaneapp
The paper introduces Brainoware as the future of AI hardware, aiming to replace artificial neural networks (ANNs).
-: FOLLOW US :- @theinsaneapp
Brainoware combines a "brain organoid" with traditional AI, creating a biohybrid computer that utilizes the decision-making power of neurons.
-: FOLLOW US :- @theinsaneapp
The lead researcher, Feng Guo, and his team cultivated clusters of specialized stem cells that evolved into neurons, the basic building blocks of the brain.
-: FOLLOW US :- @theinsaneapp
Brainoware's brain organoids demonstrate advancements in complexity, connectivity, neuroplasticity, and neurogenesis with minimal energy consumption and rapid learning capabilities.
-: FOLLOW US :- @theinsaneapp
In a speech recognition task, Brainoware achieved 78% accuracy by converting recordings of Japanese speakers into electrical pulses and training an AI to identify the speaker based on organoid neural activity.
-: FOLLOW US :- @theinsaneapp
Compared to traditional computing systems, Brainoware was less accurate in speech recognition, but it represents a significant step towards advanced biocomputing systems.
-: FOLLOW US :- @theinsaneapp
Current AI hardware using artificial neural networks consumes significantly more power (8 million watts) than a typical human brain (20 watts).
-: FOLLOW US :- @theinsaneapp
Brainoware leverages living biological neural networks within 3D brain organoids, offering a potential solution to the energy consumption and efficiency challenges of current AI hardware.
-: FOLLOW US :- @theinsaneapp
Brainoware demonstrates on-chip learning abilities via synaptic plasticity, and its potential applications include studying neurological conditions and enhancing cognitive aspects through unsupervised learning.