Is it time for a Manhattan Project in artificial intelligence?

Question

Are we in a position,ike fission in 1939, with respect to AGI? Is it possible that giving generative models a thousand-fold increase in the number of servers, memory, and improve its sensing ability beyond text scanning, and Artificial General Intelligence will appear spontaneously?

MilnerRoute · Accepted Answer

I believe the private sector is sufficiently motivated to pursue AI without the government getting involved in funding it. An atom bomb is something that has to be developed by the government (the entity that would ultimately use it). I just don't think AI is in the same category.
For what it's worth, Bill Gates disagrees with me:
"Market forces won’t naturally produce AI products and services that help the poorest. The opposite is more likely. With reliable funding and the right policies, governments and philanthropy can ensure that AIs are used to reduce inequity. Just as the world needs its brightest people focused on its biggest problems, we will need to focus the world’s best AIs on its biggest problems."
https://www.gatesnotes.com/The-Age-of-AI-Has-Begun

smoldesu · Answer

I wouldn't bet on it, there's no theory that supports this idea (to my knowledge). AGI is poorly-defined, and I feel like anything trained exclusively on text would fail to qualify.Put another way, what results are you hoping for that you're not getting out of GPT3/4 already?

spansoa · Answer

As long as AI is sand-boxed in a computer and doesn't have physical limbs, hands, legs, etc then it's safe for now. I think the real trouble starts when these things get mobile and can manipulate matter that isn't just electrons flying around a piece of silicon.

sn9 · Answer

The progress in AI isn't fast enough for you?