Hacker News new | ask | show | jobs
by elwebmaster 76 days ago
It's terrible. Fake money is fueling the exhaustion of real resources in search of questionable outcomes ("AGI"). Imagine if all of these money were invested in curing cancer.
1 comments

Imagine if AI cures cancer.
Let's also imagine an alternative reality where some reasonable percentage of the $2.5T in current year AI spending was instead invested in the "general intelligence" researchers we already have for the same purpose. I think it's a pretty reasonable expectation that 1) they'd probably make more progress and 2) that money would help a lot more people in the process (through jobs and economic activity).
You can imagine all you want, but my understanding is there is no credible evidence that scaling LLMs will result in true AGI.
Obviously there's no "evidence". Why would you even think we need AGI? But I'm happy to hear your reasoning if you were one of the few/only? people who imagined that software that could predict the next word could do what it now is doing.
An already-ageing population living even longer while nobody wants kids anymore?
Am I correctly reading your argument that you are pro cancer for the purposes of demographic balance?
I think his broader point is that life preservation doesn't seem like such a big win if overall quality of life is dropping to the point where people decide to not subject their potential children with the burden of living.
I would say that’s a pretty huge if.

Where has quality of life dropped so much that people want to die slowly at a young age of a horrible and painful disease instead of getting old?

You don't have to imagine, it will hallucinate you a slop with full confidence every time.
I've already seen at least one person who was pretty sure that the preprint paper they co-authored with AI (read: AI wrote for them) was going to cure cancer and make them billions of dollars.

There was only one problem. The paper jumped straight from "this paper will show how our new treatment could cures cancer forever" to "as you can see, these results clearly show that our treatment cures cancer" - with neither any actual results nor any specifics on the treatment. And I don't just mean that the paper didn't go into details; writing the paper was the full extent of their "research".

So QED then, I guess.
Proof by assertion.
AI was used fundamentally for COVID vaccine development. AI is used for research in all modern drugs. It’s a certainty if cancer gets cured AI will have played a fundamental role since it’s already fundamental to precursors.
Curing cancer isn't profitable. But even if someone tries to mix AI and biotech, the result will be a dangerous medical slop.
Imagine if people with inexpensive tools cure cancer. You know, like they used to so far?