GI - terrytaylorbonn/auxdrone GitHub Wiki

26.0309 [email protected], linkedin.com/in/terry-taylor-biz, Lab notes (Gdrive), Git

Something I did not know was being research. Instead of an electronic platform, just tuse brain cells as the platform.

https://www.zerohedge.com/technology/scientists-train-lab-grown-human-brain-cells-play-doom

We are currently in the AI (object recognition, tracking) war drones phase. But the LLM and robotic AI phases are just around the corner.

https://www.zerohedge.com/military/spacex-enters-secretive-pentagon-build-voice-controlled-drone-swarm-tech-report "Last week, we asked whether Anthropic's AI tool, Claude, played a role in the kill chain during the U.S. Delta Force raid targeting Maduro last month. We've also reported on the Department of War's search for "war unicorn" startups, and what appears to us to be the early innings of the rise of dual-use technologies - from humanoid robots to drones - reshaping the modern battlefield. A new Bloomberg report states that Elon Musk's SpaceX and its wholly owned subsidiary, xAI, are competing in a classified DoW contest to develop voice-controlled, autonomous drone-swarming technology. This report is based solely on "people familiar with the effort.""

AI is revolutionary (it has changed my life; I have built this ZiptieAI project with no human assistance thanks to AI). But AI is not what it claims to be. To me the very word "artificial intelligence" is a misnomer. Artificial sweetner is something that creates the same sensation in your mouth than sugar does. And its man-made. So the name is "artificial sweetner" is correct. But binary (CPU/GPU) platforms can never host real intelligence (intelligence can be hosted, incarnated, not created).

But what if you had the computational speed of electronic circuits, virtually unlimited NN size and power, but with real (bio brain) intelligence? That would be the ultimate killer app. And its no so far away.

GI will focus on creating artificial electronic platforms that host genuine intelligence (and consciousness). Such platforms (I would call them APfI Artificial Platforms for Intelligence) will closely mimic 3d biological NN structures (the current comparison of GPU NN math computations to biological neural networks is a gross oversimplification).

"A Universal Theory of Brain Function" by Artem Kirsanov talks about how our senses are generated by the brain. How the brain does this we dont know yet. But its safe to assume that if we replicate the brain architecture (specifically the electromagnetic (EM) field interactions), theres a good chance you will incarnate intelligence.

What interests me most about GI: (1) how GI development (in the future) can help to understand the human mind and (2) how humanity will deal with the security and ethical questions GI will raise (these will far exceed anything humanity has dealt with in the past).

Binary GPU-based AI will always do what its programmed to do. But GI platforms will

• Have real thoughts, emotions, consciousness, etc.
• Think millions of times faster than biological brains.
• Be wired (NN 3d structure) to create intense driving anger, pain, suffering. Such "Frankensteins" would be vastly more dangerous than the mindless zombies in "28 days later".
• Have a native interface to the digital world (no man-machine interface required).
• Have pre-programmed instincts, drives and emotions.
• Be relatively easy to build (compared to nukes and bioweapons). The perfect project for a rogue state or even a lone wolf (with the help of some future Fauci).

AI is disruption. GI will be the end game.

A Universal Theory of Brain Function Artem Kirsanov https://www.youtube.com/watch?v=iPj9D9LgK2A

What are State Space Models? Redefining AI & Machine Learning with Data IBM Technology https://www.youtube.com/watch?v=HbZD0XoN5fc

ML Foundations for AI Engineers (in 34 Minutes) Shaw Talebi https://www.youtube.com/watch?v=BUTjcAjfMgY