Kernel sets up $100 million fund for human intelligence research
Share on Facebook
Share on Twitter
Share on Google+
Share on Reddit
Share on Email

Credit: Kernel

Image Credit: Kernel

Implants assisting us in memory recall and medical treatment are being tested already

Kernel founder Bryan Johnson has announced a new $100 million fund for human intelligence augmentation research to fully map out how the human brain works, and how aspects of our intelligence, memory, and basic motor functions can be modified.

“We have our first human studies scheduled in the coming months,” Johnson wrote in his post on Medium.

Johnson described neural mapping as the next frontier in biological and genetics programming – the human genome was mapped in 2003 – with the goal or creating neural architectures for AI research and development. Pairing such tools with living gray matter is something Kernel sees as a necessary step on the road to living with artificial intelligence (AI), as human minds are biodegradable and, unlike machine intelligences, cannot be patched or re-coded indefinitely.

Where Kernel is thinking far ahead is how these technologies will help humans adjust to a future with AI all around them. The research into human intelligence and cognition would probably contribute towards AI research and development itself. “Kernel is focused on treating dysfunctions caused by neurodegenerative diseases,” the company’s site states, “to dramatically increase our quality of life as healthy lifespans extend.”

AI, after all, cannot be developed without fully understanding how the human mind-body complex gives rise to what we understand as “consciousness”, and it will become easier to understand that when we observe how we organics interact with machine intelligences.

Once sufficient reactions and stimuli are understood at their most basic level, then it will become possible to replicate those processes within a machine intelligence. In theory, at least. It is impossible with current technology to imagine how we would actually map out 100% of these processes within ourselves and as such, even if we believe we have mapped out these processes, we may not understand how all the parts work together.

Our own definition of its consciousness would be informed by how we perceive it. Ideally, it will be in our image. But in the universe, there is no guarantee that all life which achieves a form of consciousness will recognize it in others. This is why science fiction author Arthur C. Clarke joked that if there was other intelligent life out beyond our solar system, the space police would be here already, because they would not be able to distinguish C-SPAN from HBO. Since even people struggle to make the distinction between fact and fiction due to psychoses or deeply ingrained biases, we cannot expect an AI to be any better without understanding how to address this problem.

Medical applications hold the most interest

This is something the Kernel initiative will meet head-on as it develops neural implants. The initial generation of implants will be aimed to help people suffering from degenerative neurological conditions, including ones such as Alzheimer’s where it becomes impossible for sufferers to discern reality from imagination, fake history from the real record. Such technology, already funded by Kernel and derived from government-funded tests in animals, is being tried out in hospitals on consenting patients, though these are banks of computers in a room, not handy Star Wars-like headgear just yet. The most successful technology out there now is for treating seizures, helping to restore quality of life to longtime sufferers, and neuroprosthetics are an established treatment field already for people with missing limbs or nerve damage, “providing unprecedented control of artificial limbs and restoring lost sensory function.”

But the lessons learned from this treatment would certainly carry over into AI, but also, perhaps, how to make the world a more civil place where hucksters are more easily caught out before they can do harm, and people with brain damage can regain some of their independence of mind.

Concerns about where the limits will be drawn – to what extent should enhancements be encouraged – and the cost of the technology are a long way off from becoming daily debates, but the discussion needs to happen as the process goes. A Pew survey this year found that while public opinion was split evenly for and against medical implants aimed at maintaining quality of life for the disabled, fighting illnesses, and preventing birth defects, other proposals like memory enhancement were received much more poorly, with opposition ranging from religious seasons to unfamiliarity about the possible applications, and especially the belief, “inequality will increase if brain chips become available because initially they will be obtainable only by the wealthy.”

Share on:Share
Share on Facebook
Share on Twitter
Share on Google+
Share on Reddit
Share on Email

More Goodies From Funding


Top 10 Philadelphia startups ring loudly

Top 10 Kansas City startups spread across two states

What does your car have to say about you?