recked_wralph comments on Jensen Huang says Nvidia engineers should use AI tokens worth half their annual salary every year to be fully productive

[–] recked_wralph@lemmy.world 12 points 7 months ago (3 children)

Wouldn't an AI researcher naturally find generative AI disadvantageous because they are attempting to develop novel tools which could not exist in the training set in the first place?

permalink

fedilink

source

hideshow 3 child comments

[–] 8oow3291d@feddit.dk 7 points 7 months ago (2 children)

Even novel solutions are usually built out of smaller common building blocks. E.g. many novel solutions surely use a database. You can make the LLM help you set up and use the database, that your novel solution uses.

permalink

fedilink

source

parent

hideshow 2 child comments

[–] ayyy@sh.itjust.works 12 points 7 months ago

“No we don’t need databases anymore, only blockchains.” —Nvidia CEO a few years ago

permalink

fedilink

source

parent

[–] 8oow3291d@feddit.dk 0 points 7 months ago

The database was an arbitrary example. A more relevant example would be tenserflow layers in a neural network. As I understand it, you can in some cases get a novel solution to a problem just by choosing a smart enough combination, with the right data.

ChatGPT absolutely knows how to help doing the grunt work setting up the tenserflow configuration, following your directions.

permalink

fedilink

source

parent

[+] MangoCats@feddit.it 3 points 7 months ago

[removed by mod]

permalink

fedilink

source

parent

[–] 8oow3291d@feddit.dk 0 points 7 months ago (-1 children)

If used by an expect developer, then the combinations are not just random "lucky" choices.

permalink

fedilink

source

parent

hideshow -1 child comments

[+] MangoCats@feddit.it 1 point 7 months ago

[removed by mod]

permalink

fedilink

source

parent

[–] 8oow3291d@feddit.dk 1 point 7 months ago (-1 children)

The world is not that simple. There are too many combinations to try. And you risk hitting local maxima, even if doing the gradient thing.

permalink

fedilink

source

parent

hideshow -1 child comments

load more comments (-1 replies)

load more comments (-2 replies)

[–] baahb@lemmy.dbzer0.com 1 point 7 months ago*

If you are capable of giving good directions...

I'm probably not arguing with you, and I'm not trying to regardless. You seem like you have tried this, watched it happen, go "huh, neet!" And then get it to take the next step in whatever you were doing in the first place only to find out you didn't provide adequate requirements for your config.

permalink

fedilink

source

parent

[–] 8oow3291d@feddit.dk 1 point 7 months ago* (last edited 7 months ago)

Well, yes, that is a central point.

I am a senior programmer. LLMs are amazing - I know exactly what I want, and I can ask for it and review it. My productivity has gone up at least 3-fold, with no decrease in quality, by using LLMs responsibly.

But it seems to me that some people on social media just can't imagine using LLMs in this way. They just imagine that all LLM usage is vibe coding, using the output without understanding or review. Obviously you are very unlikely to create any fundamentally new solutions if you only use LLMs that way.