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

ChatGPT-4 significantly increased performance of business consultants

d3.harvard.edu

Post reply on HN
Print view View original

161–170 of 282 posts

Re: ChatGPT-4 significantly increased performance of business consultants

#161

Post by zare_st »

simonw wrote: post #149
gumballindie wrote: post #108

Earlier quoted context omitted.

> They can be worse than worthless. They can sabotage your work if you let them making you spend even more time fixing it afterwards. I basically gave up on llms because i was spending more time figuring out what it did wrong than actually getting value. People without programming skill are still impressed by them. But they yet have to learn or deliver anything of value even with the help of chat bots.

I have twenty years of programming experience and LLMs give me a significant productivity boost for programming on a daily basis: https://simonwillison.net/2023/Sep/29/llms-podcast/#code-int...

Sorry, I may have gotten something wrong by skimming over your link. Is this the "significant project" you have been assisted by LLMs?

https://github.com/simonw/sqlite-history

Re: ChatGPT-4 significantly increased performance of business consultants

#162

Post by brap »

awestroke wrote: post #5

Not surprised. It's frighteningly good, and a perfect match for programming. I often ask GPT4 to write code for something, and try if it works, but I seldom copy and paste the code it writes - I rewrite it myself to fit into the context of the codebase. But it saves me a lot of time when I am unsure about how to do something. Other times I don't like the suggestion at all, but that's useful as well, as it often clari…

I used ChatGPT yesterday for code for the first time.

I gave it a nontrivial task I couldn’t google a solution for, and wasn’t sure it was even possible:

Given a python object, give me a list of functions that received this object as an argument. I cannot modify the existing code, only how the object is structured.

It gave me a few ideas that didn’t quite work (e.g modifying the functions or wrapping them in decorators, looking at the current stack trace to find such functions) and after some back and forth it came up with hijacking the python tracer to achieve this. And it actually worked.

The crazy thing is that I don’t believe it encountered anything like this in its training set, it was able to put pieces together which is near human level. When asked, it easily explained the shortcomings of this solution (e.g interfering with the debugger).

Re: ChatGPT-4 significantly increased performance of business consultants

#163

Post by simonw »

zare_st wrote: post #161
simonw wrote: post #149

Earlier quoted context omitted.

I have twenty years of programming experience and LLMs give me a significant productivity boost for programming on a daily basis: https://simonwillison.net/2023/Sep/29/llms-podcast/#code-int...

Sorry, I may have gotten something wrong by skimming over your link. Is this the "significant project" you have been assisted by LLMs? https://github.com/simonw/sqlite-history

That's one of about a dozen at this point - but yeah, that's the one that I used LLMs to research the initial triggers and schema design for.

Here's the transcript (it pre-dates the ChatGPT share feature): https://gist.github.com/simonw/1aa4050f3f7d92b048ae414a40cdd...

I wrote more about it here: https://simonwillison.net/2023/Apr/15/sqlite-history/

Here's another one I built using AppleScript: https://github.com/dogsheep/apple-notes-to-sqlite - I wrote about that here: https://til.simonwillison.net/gpt3/chatgpt-applescript

Re: ChatGPT-4 significantly increased performance of business consultants

#164

Post by WendyTheWillow »

mlinhares wrote: post #153
djtango wrote: post #104

Earlier quoted context omitted.

Ha it's fun to dunk on management consultants but I think the magic is they are like pop music producers. Somehow they're able to make the C suite hoover up LLM shovelware the same way top producers can take super obvious music and sell I V vi IV but when we try the same chords it's uninspired and no one wants to listen to it

It’s insane that people think producing pop music is easy. Competing on the pop music market is a cutthroat business with a lot of competition and the good ones get their price back in gold. When Scorpions wanted to be resurrected they hired Desmond Child to produce them and he absolutely crushed. These people are very good at what they do and there are very few of them.

It’s also insane people think running a large business is easy, or that management consultants aren’t worth what they’re paid.

Re: ChatGPT-4 significantly increased performance of business consultants

#165

Post by baq »

qingcharles wrote: post #143
benreesman wrote: post #46

LLMs are stunningly good at language tasks: almost all of what us old-timers called NLP is just crushed these days. Summarization, Q&A, sentiment, the list goes on and on. Truly remarkable stuff. And where there isn’t a bright line around “fact”, and where it doesn’t need to come together like a Pynchon novel, the generative stuff is smoking hot: short-form fiction, opinion pieces, product copy? Massive productivity…

Try asking it for specifically NLP-ized text and it does it very well... (but then tells you not to use it as it is "unethical")

You aren't talking about the NLP the GP is talking about.

Re: ChatGPT-4 significantly increased performance of business consultants

#166

Post by aofjfdgnionio »

benreesman wrote: post #46

LLMs are stunningly good at language tasks: almost all of what us old-timers called NLP is just crushed these days. Summarization, Q&A, sentiment, the list goes on and on. Truly remarkable stuff. And where there isn’t a bright line around “fact”, and where it doesn’t need to come together like a Pynchon novel, the generative stuff is smoking hot: short-form fiction, opinion pieces, product copy? Massive productivity…

I don't buy it. LLMs cannot do anything reliably, no matter how constrained the domain. Their outputs are of acceptable quality when back to a person who will use their human brain to paper over the cracks. People can recognize when the output is garbage, figure out minor ambiguities, and subconsciously correct minor factual or logical errors. But I would never feed LLM results directly into another computer program This rules out most traditional NLP tasks.

Re: ChatGPT-4 significantly increased performance of business consultants

#167

Post by dbalatero »

vorticalbox wrote: post #123
cube00 wrote: post #54

Earlier quoted context omitted.

If your docs and PR descriptions can be generated off file diffs everyone's time could be better spent scanning the diff to come to the same conclusions. Consider using your PRs and docs to capture the answers to the usual why questions which LLM won't be able to do.

The why is largely in the ticket and the what in the pr.

Huh, your tickets aren't just a single vague title sentence and no description body?

Re: ChatGPT-4 significantly increased performance of business consultants

#169

Post by jack_riminton »

WendyTheWillow wrote: post #164
mlinhares wrote: post #153

Earlier quoted context omitted.

It’s insane that people think producing pop music is easy. Competing on the pop music market is a cutthroat business with a lot of competition and the good ones get their price back in gold. When Scorpions wanted to be resurrected they hired Desmond Child to produce them and he absolutely crushed. These people are very good at what they do and there are very few of them.

It’s also insane people think running a large business is easy, or that management consultants aren’t worth what they’re paid.

Worth here is the key word. I did an MBA which was specifically designed to get people into consulting, we did many consulting projects for real multinantionals and the key lesson was: "What does the hidden client want?"

i.e. someone has been tasked with getting some consultants to come up with a suggestion, the key question however is what does THEIR boss want to hear. If you can work that out and give it to them then you've earned your 'worth'

Re: ChatGPT-4 significantly increased performance of business consultants

#170

Post by benreesman »

aofjfdgnionio wrote: post #166
benreesman wrote: post #46

LLMs are stunningly good at language tasks: almost all of what us old-timers called NLP is just crushed these days. Summarization, Q&A, sentiment, the list goes on and on. Truly remarkable stuff. And where there isn’t a bright line around “fact”, and where it doesn’t need to come together like a Pynchon novel, the generative stuff is smoking hot: short-form fiction, opinion pieces, product copy? Massive productivity…

I don't buy it. LLMs cannot do anything reliably, no matter how constrained the domain. Their outputs are of acceptable quality when back to a person who will use their human brain to paper over the cracks. People can recognize when the output is garbage, figure out minor ambiguities, and subconsciously correct minor factual or logical errors. But I would never feed LLM results directly into another computer program…

I’m sympathetic to the instinct to push back on the absurd boosterism (these things are an existential threat to humanity this year), it’s fucking annoying.

But they can do plenty of useful stuff reliably. It’s not “be generally intelligent”, which they are just nothing even remotely close to, but know you don’t dig the LLM hype from that comment? Yeah, they get that every time.

Post reply on HN