What is a full stack AI company?
AI is making huge changes but we have not reached an intelligence explosion.
We have not transformed all transportation, labor or computer coding.
A company that can fully unlock the AI future would have control all aspects of what is needed for better and better AI.
This would be the AI chips, energy production for larger and larger AI data centers.
Those data centers would produce better and better AI models.
Those AI models would be put into products that would transform manufacturing, transportation and speed the improvement of the AI hardware and the key aspects of energy, data and compute for an intelligence, money and productivity flywheel.
A complete AI company able to make the Energy, the training and inference chips, the data centers, gather more data, make more synthetic data, and make the key products of humanoid bot, robotaxi and automated coding that address the biggest markets of replacing all labor, disrupting transportation and disrupting computer coding.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
I’ve been wondering lately just what the bottlenecks to Ai roll-out to replace human labor will be.
How much processing infrastructure is required to replace a given amount of human work? I presume work won’t all be equally easily replaced. Maybe a PC with a GPU running a tiny but specially tuned model could, fully applied and running 24/7, replace 10% of the tasks of many white collar jobs? Maybe one H100 (or 100 of them working together, timesliced over 100 tasks) could displace 50% of many WC jobs?
But then – using such estimates – how many GPUs and H100s are needed to get to that level across, say, all WC jobs in the USA? How soon are we likely to get to various levels of work displacement, assuming the hardware is the biggest bottleneck? And if that isn’t the bottleneck, what is?
One question that you never answer , who will buy those services if no one has a job to buy them?
I think the idea is that most of the population will live on UBI. Keyword: most. To me, that doesn’t bode well for most people. But there’s to know before it happens.
UBI is a double-edged sword. It will likely be needed to save us from the Ai depression as human work is displaced. But if created by government, it will be national rather than universal, and each government will certainly use it as an instrument of control over people, making most people into peasants.
The best solution I’ve found is to have a cryptocurrency that intrinsically generates the UBI into individuals’ accounts, as well as applying demurrage to all existing currency to prevent hyperinflation. Completely autonomous, globally available to anyone with a smartphone. Adoption and unique identity verification are resolvable issues.
UBI negates itself. Communism with even less incentive.
I agree, Brian.
The trick will be to build the factories quickly to keep up with AI advancements and keeping up with demand.