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

Let's be honest, Generative AI isn't going all that well

garymarcus.substack.com

Post reply on HN
Print view View original

201–210 of 346 posts

Re: Let's be honest, Generative AI isn't going all that well

#201

Post by habinero »

sheeh wrote: post #101
habinero wrote: post #66

Earlier quoted context omitted.

Every professional SWE is going to stare off into the middle distance, as they flashback to some PM or VP deciding to show everyone they still got it. The "how hard could it be" fallacy claims another!

As someone who is more involved in shaping the product direction rather than engineering what composes the product - I will readily admit many product people are utterly, utterly clueless. Most people have no clue the craftsmanship, work etc it takes to create a great product. LLMs are not going to change this, in fact they serve as a distraction. I’m not a SWE so I gain nothing by being bearish on the contributions…

Oh, it wasn't a bash on product people, I'm sorry if it came off that way.

It's a reference to a trope where the VP of Eng or CTO (who was an engineer decades ago) gets it in their head that they want to code again and writes something absolute dogshit terrible because their skills have degraded. Unfortunately they are your boss's boss's boss and can make you deal with it anyways.

I've actually seen it IRL once, to his credit the dude finally realized the engineer smiles were pained grimaces and it got quietly dropped lol.

Re: Let's be honest, Generative AI isn't going all that well

#202

Post by bonesss »

habinero wrote: post #66
fzeroracer wrote: post #33

Earlier quoted context omitted.

> Meanwhile, my cofounder is rewriting code we spent millions of salary on in the past by himself in a few weeks. This is one of those statements that would horrify any halfway competent engineer. A cowboy coder going in, seeing a bunch of code and going 'I should rewrite this' is one of the biggest liabilities to any stable system.

Every professional SWE is going to stare off into the middle distance, as they flashback to some PM or VP deciding to show everyone they still got it. The "how hard could it be" fallacy claims another!

LLMs do the jobs of developers, thereby eating up countless jobs.

LLMs do the jobs of developers without telling semi-technical arrogant MBA holders “no, you’re dumb”, thereby creating all the same jobs as before but also a butt-ton more juggling expensive cleanup mixed with ego-massaging.

We’re talking a 2-10x improvement in ‘how hard could it be?’ iterations. Consultant candy.

Re: Let's be honest, Generative AI isn't going all that well

#203

Post by sublinear »

All this AI discussion has done is reveal how naive some people are.

You're not losing your job unless you work on trivial codebases. There's a very clear pattern what those are: startups, greenfield, games, junk apps, mindless busywork that probably has an existing better tool on github, etc. Basically anything that doesn't have any concrete business requirements or legal liability.

This isn't to say those codebases will always be trivial, but good luck cleaning that up or facing the reality of having to rewrite it properly. At least you have AI to help with boilerplate. Maybe you'll learn to read docs along the way.

The people claiming to be significantly more productive are either novice programmers or optimistic for unexplained reasons they're still trying to figure out. When they want to let us know, most people still won't care because it's not even the good kind of unreasonable that brings innovation.

The only real value in modern LLMs is that natural language processing is a lot better than it used to be.

Are we done now?

Re: Let's be honest, Generative AI isn't going all that well

#204

Post by habinero »

BeetleB wrote: post #91
drewbug01 wrote: post #32

Earlier quoted context omitted.

> If your metric is an LLM that can copy/paste without alterations, and never hallucinate APIs, then yeah, you'll always be disappointed with them. I struggle to take comments like this seriously - yes, it is very reasonable to expect these magical tools to copy and paste something without alterations. How on earth is that an unreasonable ask? The whole discourse around LLMs is so utterly exhausting. If I say I don't…

For a long time, I've wanted to write a blog post on why programmers don't understand the utility of LLMs[1], whereas non-programmers easily see it. But I struggle to articulate it well. The gist is this: Programmers view computers as deterministic . They can't tolerate a tool that behaves differently from run to run. They have a very binary view of the world: If it can't satisfy this "basic" requirement, it's crap.…

> And so they get really thrown off when people use computers in a nondeterministic way to solve a deterministic problem

Ah, no. This is wildly off the mark, but I think a lot of people don't understand what SWEs actually do.

We don't get paid to write code. We get paid to solve problems. We're knowledge workers like lawyers or doctors or other engineers, meaning we're the ones making the judgement calls and making the technical decisions.

In my current job, I tell my boss what I'm going to be working on, not the other way around. That's not always true, but it's mostly true for most SWEs.

The flip side of that is I'm also held responsible. If I write ass code and deploy it to prod, it's my ass that's gonna get paged for it. If I take prod down and cause a major incident, the blame comes to me. It's not hard to come up with scenarios where your bad choices end up costing the company enormous sums of money. Millions of dollars for large companies. Fines.

So no, it has nothing to do with non-determinism lol. We deal with that all the time. (Machine learning is decades old, after all.)

It's evaluating things, weighing the benefits against the risks and failure modes, and making a judgement call that it's ass.

Re: Let's be honest, Generative AI isn't going all that well

#205

Post by mycall »

array_key_first wrote: post #200
jbgt wrote: post #191

Earlier quoted context omitted.

Slice THEIR hands. They might say yours are rigged. I'm a non dev and the things I'm building blow me away. I think many of these people criticizing are perhaps more on the execution side and have a legitimate craft they are protecting. If you're more on the managerial side, and I'd say a trusting manager not a show me your work kind, then you're more likely to be open and results oriented.

From a developer POV, or at least my developer POV, less code is always better. The best code is no code at all. I think getting results can be very easy, at first. But I force myself to not just spit out code, because I've been burned so, so, so many times by that. As software grows, the complexity explodes. It's not linear like the growth of the software itself, it feels exponential. Adding one feature takes 100x t…

You can use AI to simplify software stacks too, only your imagination limits you. How do you see things working with many less abstraction layers?

I remember coding BASIC with POKE/PEEK assembly inside it, same with Turbo Pascal with assembly (C/C++ has similar extern abilities). Perhaps you want no more web or UI (TUI?). Once you imagine what you are looking for, you can label it and go from there.

Re: Let's be honest, Generative AI isn't going all that well

#206

Post by Gud »

vips7L wrote: post #171
maccard wrote: post #113

Earlier quoted context omitted.

Someone else said this perfectly farther down: > The whole discourse around LLMs is so utterly exhausting. If I say I don't like them for almost any reason, I'm a luddite. If I complain about their shortcomings, I'm just using it wrong. If I try and use it the "right" way and it still gets extremely basic things wrong, then my expectations are too high. As I’ve said, I use LLMs, and I use tools that are assisted by L…

Rather be a Luddite than contribute to these soul suckers like OpenAI and help them lay off workers.

How are they “soul suckers”?

Using LLMs has made it fun for me to make software again.

Re: Let's be honest, Generative AI isn't going all that well

#208

Post by suddenlybananas »

retrac wrote: post #131
daveguy wrote: post #129

Earlier quoted context omitted.

"quite obscure" doesn't mean there is nothing in the internet that directly addresses the question. Here is an english analysis of the text that easily showed up in an internet search: https://www.cantab.net/users/leonardo/Downloads/Varian%20Sym... This source includes analysis of "the Black Stone."

Not quite the same analysis. The human is better, no surprise. But the NotebookLM output links back to the original book in a very useful way. If you think about it as fuzzy semantic search it's amazing. If you want an essay or even just creativity, yes it's lacking.

Surely there is analysis available online in French though?

Re: Let's be honest, Generative AI isn't going all that well

#209

Post by suddenlybananas »

dyauspitr wrote: post #152
gejose wrote: post #42

I believe Gary Marcus is quite well known for terrible AI predictions. He's not in any way an expert in the field. Some of his predictions from 2022 [1] > In 2029, AI will not be able to watch a movie and tell you accurately what is going on (what I called the comprehension challenge in The New Yorker, in 2014). Who are the characters? What are their conflicts and motivations? etc. > In 2029, AI will not be able to r…

In my opinion, contrary to other comments here I think AI can do all of the above already except being a kitchen cook. Just earlier today I asked it to give me a summary of a show I was watching until a particular episode in a particular season without spoiling the rest of it and it did a great job.

You know that almost every show as summaries of episodes available online?

Re: Let's be honest, Generative AI isn't going all that well

#210

Post by suddenlybananas »

nullpoint420 wrote: post #187
daxfohl wrote: post #142

Earlier quoted context omitted.

Though I think it's a very steep sigmoid that we're still far on the bottom half of. For math it just did its first "almost independent" Erdos problem. In a couple months it'll probably do another, then maybe one each month for a while, then one morning we'll wake up and find whoom it solved 20 overnight and is spitting them out by the hour. For software it's been "curiosity ... curiosity ... curiosity ... occasional…

And then what? Am I supposed to be excited about this future?

You have to remember that half these people think they are building god.
Post reply on HN