[imagesource:gencraftai]
The AI revolution is speeding ahead, and it’s taking our brains with it.
That is a bit of a double entendre, as we come to rely more and more on smart software like ChatGPT to formulate ideas and do our work while artificial intelligence is also busy extending into something called ‘Brainoware’.
To explain the latter situation, biotech enthusiasts have connected clusters of human brain cells to a computer, allowing it to perform a rudimentary form of speech recognition. The ultimate aim, they say, is to create systems that consume significantly less energy for AI tasks compared to traditional silicon chips.
This may just make the ‘A’ in AI obsolete.
A research paper formally published today in the journal Nature Electronics featured work by scientists at Indiana University Bloomington, explaining the new ‘Brainoware’ system that they say uses human brain organoids in completing advanced AI tasks.
Brain organoids, essentially clusters of nerve cells resulting from the growth of stem cells in specific conditions and which resemble organs, are described by one of the study’s co-authors, Feng Guo as “mini-brains.” Since organoids are made from stem cells capable of specialising into different types of cells, they were morphed into neurons, akin to those found in our brains, in this case.
Both AI and the brain rely on passing signals around a web of interconnected nodes, known as a neural network. “We wanted to ask the question of whether we can leverage the biological neural network within the brain organoid for computing,” he says.
Creating the Brainoware system involves positioning a solitary organoid onto a plate equipped with thousands of electrodes, establishing connections between the brain and electric circuits. Subsequently, the researchers encode the desired input information into a series of electric pulses, which are then transmitted to the organoid. The sensor captures the brain tissue’s response, and this information is “decoded” through a machine-learning algorithm capable of discerning the associated information.
This is part of a brain organoid, in which stem cells (pink) are differentiating into neurons (purple):
To evaluate the capabilities of Brainoware, the team employed the method of voice recognition. They trained the system using 240 recordings of eight individuals speaking, converting the audio into electric signals transmitted to the organoid. The miniature brain exhibited distinct reactions to each voice, resulting in varied patterns of neural activity. The AI successfully acquired the ability to interpret these responses, enabling it to identify the speaker. Post-training, the system achieved a voice identification accuracy of 78%.
The organoids are still quite primitive today. However, researchers hope their use paves the way for biocomputers that can execute the same tasks that computers do but with minimum energy consumption.
“A human brain typically expends about 20 watts, whereas current AI hardware consumes about 8 million watts, to drive a comparative ANN (artificial neural network),” the research paper argues. “Brainoware could provide additional insights for AI computing because brain organoids can provide BNNs (biological neural networks) with complexity, connectivity, neuroplasticity, and neurogenesis, as well as low energy consumption and fast learning.”
Michael Le Page noted that “Human brains use far less energy and learn much faster, so some researchers see biocomputing as the way forward,” on Twitter in March, but noted that pushing the field to the limit could raise thorny questions.
Le Page quoted Cambridge developmental neurobiologist Madeline Lancaster, who said, “Whether these push them beyond an ethical limit is something we certainly want to avoid, and the scientific and ethics community is coming together to define where that limit would be.”
Feng Guo from Indiana University Bloomington explains, “This serves as a proof-of-concept to demonstrate our capability, but we acknowledge there’s a considerable journey ahead.”
Did you ever think you would live to see the day when ‘brain’, ‘AI’ and ‘organoids’ are used together in a sentence, let alone to create new technology? What a time to be alive.
[sources:newscientist&nature]
