Hacker News Reader

Comment by atleastoptimal | Hacker News Reader

I know it sucks now and I agree GPT-4 is not a replacement for coders. However the leap between GPT-3 and 4 indicates that by the 6 level, if improvements continue, it'll reach the scope and accuracy we expect from highly paid skilled humans.

It's only a guess people make that AI improvements will stop at some arbitrary point, and since that point seems to always be a few steps down from the skill level of the person making that prediction, I feel there's a bit of bias and ego driven insecurity in those predictions.

View on HN

> However the leap between GPT-3 and 4 indicates that by the 6 level, if improvements continue, it'll reach the scope and accuracy we expect from highly paid skilled humans.

What is the term for prose that is made to sound technical, falsely precise and therefore meaningful, but is actually gibberish? It is escaping me. I suppose even GPT 3.5 could answer this question, but I am not worried about my job.

Fundamentally it cannot reach the scope or accuracy of a highly skilled person. It's a limitation of how LLMs function.

Do you honestly think no AI advancement will fix those limitations? That LLM's or their successors will just never reach human level no matter how much compute or data are thrown at them?

No, we won't. Not in either of our lifetimes. There are problems with infinitely smaller problem spaces that we cannot solve because of the sheer difficulty of the problem. LLMs are the equivalent of a brute force attempt at cracking language models. Language is an infinitesimal fraction of the whole body of work devoted to AI.

Ok. Check back on this thread in 3 years then.

Done, see you in three years.

comment time limit is 14 days, not sure if you can keep it alive for 3 years by commenting 160 deep

They could create a new post, resurfacing this bet.

how will the other person ever find it

They could … share email addresses.

You should really make a bet on longbets.org if you're serious.

That's what they used to say about Go before DeepMind took Lee Se-dol for a ride.

Not bad for a parrot.

As for language, LLMs showed that we didn't really understand what language was. Don't sell language short as a concept. It does more than we think.

>> Do you honestly think no AI advancement will fix those limitations? That LLM's or their successors will just never reach human level no matter how much compute or data are thrown at them?

It has not happened yet.

If it does, how trustworthy would it be? What would it be used for?

HAL-9000 (https://en.wikipedia.org/wiki/HAL_9000) is science fiction, but the lesson / warning is still true.

In terms of scope, it's already left the most highly-skilled people a light year behind. How broad would your knowledge base be if you'd read -- and memorized! -- every book on your shelf?

plausible, but also i think a highly paid skilled person will do a lot worse if not allowed to test their code, run a compiler or linter, or consult the reference manual, so gpt-4 can get a lot more effective at this even without getting any smarter

Reply on news.ycombinator.com