CL1 – Synthetic Biological Intelligence
Some time ago I posted about a ‘wetware’ neural network that could be taught to play Pong. Now that early breakthrough has been refined and commercialised as the CL1 by Cortical Labs. Cortical Labs is a research and development company based in Melbourne, Australia. I’m both proud and chilled.
I’m proud because, hey, Cortical Labs in an Aussie company and it’s based in my home town. The chilled part is a little harder to explain, so bear with me. AI as it’s currently defined is ‘simply’ a neural network that’s base on hardware and software, all of it digital. Digital is great for logic, but it doesn’t even come close to replicating what we mean when we talk about ‘sentience’.
As humans, we cannot avoid thinking of sentience as the way we think. But the way we think is a curious mixture of both electrical stimulus and chemical pathways.
To give you a very, VERY simplistic example, our nerve endings register a stimulus – pin prick, hot stove, freezing cold etc – and each one sends that stimulus along electrical pathways until they reach a synapse.
Think of the synapse as a lake. To cross that lake you need ferries. That’s where the chemical part comes in. The lake is full of chemicals [hormones]. Depending on how strong the electrical stimuli are, they will be translated into the appropriate chemicals. Again, the strength of the chemical reaction will depend on the strength of the electrical stimuli that triggered them.
The lake analogy is not a good one but it’s the best I could do. Anyway, on the other ‘side’ of the lake, the chemical translations are once again changed into electrical stimuli and that is how they continue their journey…millions and millions of translations across millions and millions of synapses. Some get stronger, some get weaker, but eventually all those electrical>chemical>electrical processes trigger a response[s] from higher level parts of the human brain.
In humans, a pain response – ouch! that hurt – teaches us to stay away from hot stoves etc. By contrast, something that triggers a dopamine response – oooh, that feels nice – will make us want to do that ‘thing’ again and again. Because it feels good.
This is the ‘carrot and the stick’ that is at the heart of much human behaviour. But these behaviours are all analogue, meaning there are almost infinite shades of difference leading to almost infinite changes in learned behaviour. This is part of what makes each one of us unique.
I once read that a single human brain is the equivalent of something like 17 billion computers. Admittedly that was quite a few years ago, but the point still stands. More importantly, no matter how large LLMs – Large Language Models – like ChatGPT become, they can never become sentient like us because they are based on just the digital half of the equation.
So what of CL1?
Cortical Labs does not use brain cells per se. Instead, they use “human-induced pluripotent stem cells (hiPSCs) integrated into high-density multielectrode arrays (HD-MEAs) based on complementary metal–oxide–semiconductor (CMOS) technology could be electro-physiologically stimulated to forge autonomous, highly efficient information-exchange paths.” New Atlas
By creating their own brain cells, Cortical Labs gets around the ethical problem of using cells from someone who was once alive, i.e. a real person. But, if this wetware continues to evolve by responding to the equivalent of the ‘pleasure pain’ stimuli, won’t it eventually evolve some sense of ‘self’? After all, these pluripotent stem cells do contain DNA, and that DNA could dictate how they ‘evolve’. Have we just taken the first step in creating a cyborg?
I don’t know the answer to that question. No one does, but I’m chilled at the thought that we could be on the threshold of creating an artificial brain that is closer to us than we’d like to acknowledge, but also very, very different.
I strongly recommend that you read the entire New Atlas article which you can find here.
On a selfish note, I loved reading about a technology that creates a synthesis between wetware and digital computing. That synthesis is at the heart of the tech in Innerscape, and I imagined it over ten years ago. Yay me. Sadly, I didn’t imagine that the AI of that time would be the recipients of this wetware. This myopia is the bane of all scifi writers. We never get it all right. 
Cheers,
Meeks
