Leaders in Pharmaceutical Business Intelligence (LPBI) Group

Future of Big Data for Societal Transformation

Larry H. Bernstein, MD, FCAP, Curator

LPBI

Musk, others commit $1 billion to non-profit AI research company to ‘benefit humanity’

Spot on Elon! The threat is and currently the developments are unfortunately pointing exactly in that direction that AI will be controlled via a handful of big and powerful cooperation . None surprisingly none of those subjects are part of the OpenAI movement.

I like the sentiment, AI for all and for the common good, and at one level it seems doable but at another level it seems problematic on the scale of nation states and multinational entities.

If we all have AI systems then it will be those with control of the most energy to run their AI who will have the most influence, and that could be a “Dr. Evil”. It is the sum total of computing power on any given side of a conflict that will determine the outcome, if AI is a significant factor at all.

We could see bigger players looking at strategic questions such as, do they act now, or wait and put more resources into advancing the power of their AI so that they have better odds later, but at the risk of falling to a preemptive attack. Given this sort of thing I don’t see that AI will be a game changer, a leveller, rather it could just fit into the existing arms race type scenarios, at least until one group crosses a singularity threshold and then accelerates away from the pack while holding everyone else back so that they cannot catch up.

Not matter how I look at it I always see the scenarios running in the opposite direction to diversity, toward a singular dominant entity that “roots” all the other AI, sensor and actuator systems and then assimilates them.

How do they plan to stop this? How can one group of AIs have an ethical framework that allows them to “keep down” another group or single AI so that it does not get into a position to dominate them? How will this be any less messy than how the human super-powers have interacted in the last century?

I recommend the book “SuperIntelligence” by Nick Bostrom. Most thorough and penetrating. It covers many permutations of the intelligence explosion. The Allegory at the beginning is worth the price alone.

Elon, for goodness sake, focus! Get the big battery factory working, get space industry off the ground and America back in the ISS resupply and re-crew business, but enough with the non-profit expenditures already! Keep sinking your capital into non profits like the Hyperlink-a beautiful, high tech version of the old “I just know I can make trains profitable again outside of the northeast” dream and this non-profit AI and you’ll eventually go one financial step too far.

Both for you and for all of us who benefit from your efforts, consider this. At least change your attitude about profit; keep the option open that this AI will bring some profit, even with the open source aspect. This is a great effort, as I see you possibly becoming the “good AI” element that Ray writes about in his first essay, in the essay section on this site. There, Ray is confident that the good people with AI will out-think the bad people with AI and so good AI will prevail.