Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What it feels like to work in AI right now

robotic.substack.com

Post reply on HN
Print view View original

431–440 of 768 posts

Re: What it feels like to work in AI right now

#431

Post by jack_pp »

auggierose wrote: post #399
bottlepalm wrote: post #311

Earlier quoted context omitted.

Reading and validating code is way easier than figuring out how to write it in the first place. A lot of programming is the same - functions, conditionals, loops, libraries, etc.. The hard part is mostly semantics across languages. Way easier for GPT to do the first pass, then go through it and read up on the parts you don't understand.

That's just not true. I expect software to become a magnitude shittier soon because of this attitude. After that I expect software to become unbelievably good, because thanks to AI we will have the ability to prove correctness of software much more cheaply than before, and to design on a higher level than before. After all, you don't want the AI to help you generate boilerplate code, you want the AI to help you avoid…

It is true if you can understand the code it writes. Maybe you are worried junior programmers are gonna churn out code using ChatGPT but to be honest I'd rather trust code from ChatGPT than from a junior, I feel ChatGPT writes better code on average.

Re: What it feels like to work in AI right now

#432

Post by literalAardvark »

glandium wrote: post #403
roer wrote: post #364

Earlier quoted context omitted.

Are you using GPT4 or 3?

GPT-3. I assumed that's what most people mean with ChatGPT. Usually, people specify for 3.5 or 4. Is GPT-4 significantly better with Japanese?

GPT-4 is dramatically better with everything except speed. The minutia most people in this thread are complaining about is almost completely solved by GPT-4.

For example I used it over like 30 minutes to conceptualize, solve, and write some code that draws a graphic for a simple physics problem (0-shot) that I could intuitively understand but had no (physics and math knowledge) tools to calculate properly and it was a great experience.

It let me pick and change how what properties i wanted represented on the graph, knew what center of mass means, how to calculate it for a weird object, got something that feels correct to me, drew it out with various representations where i used color, size, shape to represent the various distances, weights, clusters, etc.

My non programmer friends have used generative networks to make designs for prints, incredible and generally accurate folk art, full mobile games.

People are sleeping on this. Don't be them.

Re: What it feels like to work in AI right now

#433

Post by throwaway2037 »

petesergeant wrote: post #376
version_five wrote: post #288

Earlier quoted context omitted.

It's a hype bubble. There's a small group if die-hards in the tech-adjacent VC community that talk it up to no end and attack anyone who disagrees (I just saw a guy ask you if you were "virtue signaling" with your post - lol. And there's lots of people exploring how it could potentially be commercially useful - just like with blockchain. As everyone points out, it's got more going for it than blockchain did because i…

I think you're wrong. I never bought into the blockchain hype, and blockchain added literally nothing to my life other than some speculation right at the beginning. I would pay $200 a month for ChatGPT, today, without thinking twice about it. That alone makes me think it's a much, much bigger deal than you think it is.

Why is this downvoted? It is a good point. My flatmate has a PhD in comp sci, does AI/ML research. He is now paying for ChatGPT "pro" because he is non-native in English and needs a bit of help to improve when writing his papers. He said ChatGPT is so good that he is learning a lot about English by just reading the corrections/improvements. And yes, we have talked/agreed about the scam/sham of blockchain/defi.

Re: What it feels like to work in AI right now

#434

Post by quickthrower2 »

amrocha wrote: post #258
agentultra wrote: post #137

Earlier quoted context omitted.

The reward for increased productivity is more work! The problem for the capitalist is having to pay all those pesky people to do the work. Eats into profits. Now we can burn out 2 people and do the work that used to take 10!

But wait, that means there 8 unemployed people without money to spend on our business... Oh I know! Get the government to give everyone money! Just not so much money that they can compete with us!

This but un-ironically! Politically it is hard ti support UBI. But print a $10k stipend for every person because the economy is tanking is obviously palatable. It is always seen as fine if you fix a problem in retrospect and without a tax rise. Money will eventually be like Robux.

Re: What it feels like to work in AI right now

#435

Post by greeny7373 »

atq2119 wrote: post #394
chaxor wrote: post #350

Earlier quoted context omitted.

There are plenty of tools that make it possible, such as langchain and the rest

Sure, quite likely - but only with significant engineering effort. I said as much in my comment. The point is that it's not a turnkey solution today , but some of the comments on here make it sound like it is. That's a form of hype.

You describe it in a way that sounds like hype is negative.

Hype can also just mean a lot of media attention.

This doesn't indicate anything good or bad.

Re: What it feels like to work in AI right now

#436

Post by AussieWog93 »

snowe2010 wrote: post #412
AussieWog93 wrote: post #343

Earlier quoted context omitted.

I've not used GPT-4, so it could be different, but regular old GPT-3.5 gets a _lot_ of things wrong.

GPT 4 is quite astounding. It might be wrong on occasion, but it will easily point you in the right direction most of the time. It still messes up, but like a twentieth of what 3.5 did. Honestly it is like an incredible rubber ducky for me. Not only can I just talk like I’m talking to a rubber duck but I can get fast, mostly informed, feedback that unblocks me. If I have a bunch of things competing for my attention I…

Is a 95% reduction in errors an exaggeration or is it really that much better? Might just need to drop the $20/mo if it's really improved that much.

Re: What it feels like to work in AI right now

#437

Post by rlt »

version_five wrote: post #375
donkeyd wrote: post #340

Earlier quoted context omitted.

> What am I missing? I'm going to guess (assume) you probably haven't worked in a 'real' business. A place where elbow grease still does the majority of the work and where Windows 8 was only just phased out. The killer app (to me) in case of GPT is GPT itself, not ChatGPT. ChatGPT just allows me to easily test use cases for GPT. There are many interesting use cases for those elbow grease businesses for GPT. For examp…

Right, questioning the value of a chatbot is a good indication someone hasn't had real job. Totally legit characterization that doesn't make the whole thing sound like a confidence game.

> Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith.

https://news.ycombinator.com/newsguidelines.html

> questioning the value of a chatbot

Original commenter was questioning the value of the entire field.

> hasn’t had a real job”

This is different than “haven't worked in a 'real' business”.

Re: What it feels like to work in AI right now

#438

Post by throwaway2037 »

have_faith wrote: post #391
fomine3 wrote: post #326

Earlier quoted context omitted.

Yo, check this out! DeepL is like way better than Google Translate, so you should totally use it. But, GPT-4 is like a game-changer, dude! It's obvi way better than DeepL, and it can do all sorts of stuff like rewording, explaining in detail, and changing up the style. This text was translated from Japanese to gamer-style English by ChatGPT-4, no joke!

This passes for “gamer-style” English?

I had to read GP twice, plus your comment to realise it might been generated by a GPT-ish service. Damn, this stuff is getting good.

Re: What it feels like to work in AI right now

#439

Post by greeny7373 »

bmitc wrote: post #207
nr2x wrote: post #158

Earlier quoted context omitted.

Natural language is now a fully functional user interface out of the box. This is bigger than the mouse.

Fully functional in what way? As far as I can tell, ChatGPT is a box that I put sentences into, and I get grammatically correct sentences that contains topics or words loosely statistically correlated to what were in my sentences, that may or may not be correct and often are not. The box has little to no memory. I honestly don't see what's so useful about this box.

You can teach it to act as a support personal.

'only answer the following questions and use these API endpoints doing the task'.

Now you have a multi lingual human Isabel Interface.

Have you seen the Wolfram alpha examples?

Re: What it feels like to work in AI right now

#440

Post by adalacelove »

hutzlibu wrote: post #369
bionhoward wrote: post #275

Earlier quoted context omitted.

It’s a hype bubble for hundreds of years and saying that doesn’t make chatgpt worth any less. I have definitely been surprised by this and gotta say I’m expecting AGI a lot faster now. Even if literally all it did was predict what the average internet user would write in a certain context, that’s huge, cuz when you integrate all the little advantages of all the weird things one person knows another doesn’t, the colle…

"Who’s gonna trust a human doctor to diagnose their stuff if the human brain holds 9 things and the AI holds thousands?" The human brain can hold much more than 9 things and even though AI will be used in medicine broadly very soon, I really want the final diagnosis done by a human. Once true AGI arrieves, I might change my opinion, but that might take a while.

> AI will be used in medicine broadly very soon

We have been hearing this since forever.

Revolutions do happen but not the way we expect. My anecdotical experience: no one in my team of about 30 people developing SW uses ChatGPT or similar in their day to day. This may change, or not.

Post reply on HN