What happens if there are no more humans in the loop, and instead of building on human work, models have to build on their own work? Would we get stuck in a local maxima?
When AI labs talk about “long-horizon tasks”, they mean tasks on the order of a few days. But what about keeping coherence across decades, like humans can? Models are not trained or tested on the ability to do that, and it seems unlikely that capability would naturally exist.
I'm pretty sure that if you put a few LLMs, agents, or anything into a room together without outside access, and get them to learn off of each other, you'll end up with gibberish, nonsensical garbage and definitely nothing magical will happen. Delirium, oblivion, dementia and so on...
Unfortunately that will never happen because the people who stand to benefit most from AI already have cost-efficient (see grossly underpaid) humans to do this for them.
They need AI to replace the expensive not cost-effective humans.
AI can do anything that it was trained on based on humans. As soon as humans are not involved, AI will have nothing to train on, and hence, the rate of improvement will approch zero, and the AI will collapse.
Human knowledge races up to the frontier through formal education and then nudges past the frontier a little bit at a time and AI will train on AI outputs that do the same. But, it will do it 1000x faster than humans because that's just how they work.
Some of the process of expanding the frontier is not parallelizable and is built on top of previous discoveries. And pushing the frontier requires doing things in the real world, or running experiments that take significant real world time that can hurt your 1000x faster claim.
I think what you are missing is that it will dramatically lower the cost of living so that everyone’s standard of living increases. It’s not about jobs.
> I think what you are missing is that it will dramatically lower the cost of living so that everyone’s standard of living increases. It’s not about jobs
This is flat out wrong. The problem is excessive taxation and government spending beyond its means that lowers everyone's standard of living.
> the people who are grasping for some indication that AI can’t do everything have become increasingly desperate. If you push hard enough, the answer is inevitably something like “the AI simply cannot [dream, love, laugh],”
Dogshit strawman article. I am willing to concede that a token predictor is better than you at whatever task you can dream up. It is not better than me, and will never be better than me, at my profession.
What happens if there are no more humans in the loop, and instead of building on human work, models have to build on their own work? Would we get stuck in a local maxima?
When AI labs talk about “long-horizon tasks”, they mean tasks on the order of a few days. But what about keeping coherence across decades, like humans can? Models are not trained or tested on the ability to do that, and it seems unlikely that capability would naturally exist.
I'm pretty sure that if you put a few LLMs, agents, or anything into a room together without outside access, and get them to learn off of each other, you'll end up with gibberish, nonsensical garbage and definitely nothing magical will happen. Delirium, oblivion, dementia and so on...
Coherence is a given... What we lose over decades is alignment.
I wish AI could do my chores. Laundry, folding clothes, dishes, cleaning... That would be nice.
If it does that, you should hope it can also stand in bread lines for you too.
Unfortunately that will never happen because the people who stand to benefit most from AI already have cost-efficient (see grossly underpaid) humans to do this for them.
They need AI to replace the expensive not cost-effective humans.
> I wish AI could do my chores. Laundry, folding clothes, dishes, cleaning... That would be nice.
Sorry, AI only does the fun stuff.
Off to the silica sand mines with you and the rest of the meat bags to keep the AI well fed.
Washing machines, dish washers, power washers, etc exist.
one day soon, right?
literally the only thing i want out of ai ^^^^
that said, it is fun generating code with them
this will of course happen in a few years...
Won’t matter. Most of us won’t have paychecks to buy it with.
Kidding (…kinda).
AI can do anything that it was trained on based on humans. As soon as humans are not involved, AI will have nothing to train on, and hence, the rate of improvement will approch zero, and the AI will collapse.
Human knowledge races up to the frontier through formal education and then nudges past the frontier a little bit at a time and AI will train on AI outputs that do the same. But, it will do it 1000x faster than humans because that's just how they work.
>1000x faster than humans
Some of the process of expanding the frontier is not parallelizable and is built on top of previous discoveries. And pushing the frontier requires doing things in the real world, or running experiments that take significant real world time that can hurt your 1000x faster claim.
what a dumb thing to keep repeating at this point
Why?
Just ignoring all the recently solved math problems human, huh?
If you understood how they were solved, you'd see that you're only bolstering the point you were responding to.
I think what you are missing is that it will dramatically lower the cost of living so that everyone’s standard of living increases. It’s not about jobs.
I don't think there's any particularly good evidence of this other than blind optimism.
that's not how it works. you get better tech for the same or more work and you can work longer before you die.
if you are no longer needed, you won't be allowed to live.
makes me chuckle every time folks think humans who get to power are nice.
Housing and healthcare are going to get cheaper??
> I think what you are missing is that it will dramatically lower the cost of living so that everyone’s standard of living increases. It’s not about jobs
This is flat out wrong. The problem is excessive taxation and government spending beyond its means that lowers everyone's standard of living.
Not always that you see two wildly wrong claims made in ostensible opposition to each other
> the people who are grasping for some indication that AI can’t do everything have become increasingly desperate. If you push hard enough, the answer is inevitably something like “the AI simply cannot [dream, love, laugh],”
Dogshit strawman article. I am willing to concede that a token predictor is better than you at whatever task you can dream up. It is not better than me, and will never be better than me, at my profession.
New is never normal.
I stopped reading when the author became overwhelmed by their compulsion to blame everything in the universe to the current US president.