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

Ask HN: Is anyone else bearish on OpenAI?

news.ycombinator.com

Post reply on HN
Print view View original

181–190 of 321 posts

Re: Ask HN: Is anyone else bearish on OpenAI?

#181

Post by austin-cheney »

I whole heartedly agree.

* Problem statement *

The actual value of GPT is spontaneous creation of spoofed data. Some of that output answers really tough questions. Stop thinking at this point and reflect.

* Value assessment *

There is some amazing potential there for performing test automation where either large data samples are demanded or the sophistication of output exceeds exceeds capabilities of prior conventions. Taken one step further there is further value in using this authentic looking out for testing against humans for market acceptance testing or bias validation.

* Real world use *

When real world use significantly departs from the value assessment there is a problem. Faking student papers or hoping the technology writes a legal brief for you forces a complete realignment of potential revenue model with different clients paying different amounts.

* Expectation *

Unless this technology can actually make money in a reoccurring and sustainable way past initial trends it will be an investment black hole just like crypto.

Re: Ask HN: Is anyone else bearish on OpenAI?

#182

Post by danielbln »

collyw wrote: post #158
ghshephard wrote: post #9

OpenAI, at least in my day-day workflow for the last 9+ months has so superseded anything that google ever was to me that I'm having a difficult time comparing the two. I've got a monitor dedicated 100% of the time to ChatGPT, and I interact with it non stop during the flow of technical scenarios and troubleshooting situations that flow into me - working in areas that I have the slimmest of backgrounds in, and shutti…

Have you checked the results it has given you? I tried it with some science related stuff. It literally made up a paper that didn't exist. I pointed this out so it linked to a real paper on an unrelated subject. For coding related stuff it is like a better stack overflow, but again it doesn't work well with niche libraries, or even some fairly big ones.

Yea, LLMs hallucinate. They almost never hallucinate if you provide the right context, either via a search enabled LLM (ChatGPT with search, or perplexity.ai or phind.com etc.) or by providing it with more code/text/input to work with. Don't use LLMs as fact database lookup machines.

Re: Ask HN: Is anyone else bearish on OpenAI?

#183

Post by Rastonbury »

smrq wrote: post #177
Rastonbury wrote: post #138

Earlier quoted context omitted.

As a non-professional coder, I'm honestly surprised there seems quite a few professional coders in this thread who don't do the first Stackoverflow first scenario. As someone learning code, shoehorning SO code when I don't understand 20% of the replies made the learning curve so steep, chatGPT has made this so much easier by changing it and telling me why will/won't work, even when watching out for or correcting hall…

As I've matured in my career I've gone to SO less and less, to the point where I mostly avoid it now. I use tools, languages, and libraries with which I am now highly familiar--perhaps more so than the average SO answerer, although I guess it would be arrogant to say for sure--and so the bulk of my questions are edge cases in my knowledge which are more easily answered by reading source code or specs. I suspect this…

That makes sense, and its funny because the other reply I got a joke was about how bad coders don't even bother going to SO

Re: Ask HN: Is anyone else bearish on OpenAI?

#184

Post by kubiton »

LLM are the first interface I as.a.software engineer is impressed by.

My company also got it immediately and is rolling it out globally.

GitHub copilot is already helpful. GitHub copilot for docs (announced at GitHub next) is a game changer.

I used openai to reformulate emails and suddenly got positive feedback about my announcement emails.

I communicate with openai German and English how ever I see fit.

It's very hard NOT to see it than the other way around.

And we got so far with only one company pushing this!

There is no option for the others to alsolo out tons of money into ai.

And besides openai, ai/ml is huge in what Nvidia and others are doing with it. Texture compression, 2d and 3d generation.

What we see is also potentially something like a new operating system.

And it makes it so much more accessible.

I never had a tool like llms which are able to take a bad copy pasta of a pdf and pulling out facts from it.

Re: Ask HN: Is anyone else bearish on OpenAI?

#185

Post by kubiton »

bmitc wrote: post #36
ghshephard wrote: post #9

OpenAI, at least in my day-day workflow for the last 9+ months has so superseded anything that google ever was to me that I'm having a difficult time comparing the two. I've got a monitor dedicated 100% of the time to ChatGPT, and I interact with it non stop during the flow of technical scenarios and troubleshooting situations that flow into me - working in areas that I have the slimmest of backgrounds in, and shutti…

Are you able to give a specific example of a problem it helped you solve? Especially one that you were at a complete blocking point, it provided some solution, and then you were able to continue to expand upon that solution. I keep reading responses like yours, but I haven't seen any specific examples of problems being solved, so it all sounds very abstract. In my interactions with ChatGPT, it felt like just interact…

I used it yesterday to generate a basic layout with image and input field based on tailwindcss.

Would have taken me much more time to figure it out than what chatgpt did at first try.

Re: Ask HN: Is anyone else bearish on OpenAI?

#186

Post by softwaredoug »

OK I'm not bearish. I'm very bullish. I use it all the time, in place of Stackoverflow type questions.

BUT

I think people are too bearish on boring-old search as a tool. It's so easy to jam a search into Chrome's bar and look for a quick reference or (hopefully) a human being that has had some experience with what you're working on.

I use search / ChatGPT / CoPilot interchangeably for different reasons... ChatGPT for a detailed, thoughtful answer. CoPilot as autocomplete on steroids. Search for reference, quick answers, and direct human experience.

Re: Ask HN: Is anyone else bearish on OpenAI?

#187

Post by danielbln »

marginalia_nu wrote: post #174
AussieWog93 wrote: post #106

Earlier quoted context omitted.

GPT-4 or GPT-3.5? They're completely different products.

We need to stop propagating this nonsense. GPT-4 still messes up.

OP said they're different products, which is very true. Yes, GPT-4 still makes mistakes, but it's leagues ahead of 3.5.

Re: Ask HN: Is anyone else bearish on OpenAI?

#188

Post by greenie_beans »

sgarland wrote: post #101
a_square_peg wrote: post #21

ChatGPT has already saved me from hours of Googling when I'm trying to find out how to do certain things. It almost feels magical - I don't have to read through half-dozen slightly different variations of what I need to do. Before ChatGPT, to find the answer to things like "how do I set up Gunicorn to run as a daemon that restarts when it fails" I would have to endure hours of googling, snarky stack-overflow comments…

I know this is just an example, but I think it’s emblematic of the main issue I have with the widespread use of LLMs. Do you mean, “have Gunicorn keep N workers running?” If so, that’s in the manual (timeouts to kill silent workers, which defaults to 30 seconds). Or do you mean “have Gunicorn itself be monitored for health, and restarted as necessary?” There are many ways to do that – systemctl, orchestration platfor…

read this

> snarky stack-overflow comments that I shouldn't do that,

and realized they're probably using the chatgpt equivalent: nice, corporate answers that you probably shouldn't use

Re: Ask HN: Is anyone else bearish on OpenAI?

#189

Post by yunwal »

OJFord wrote: post #20

When I've tried to use it myself for verifiable things (i.e. code basically) I've had the 'confidently wrong' experience. When I've seen colleagues visibly use it (i.e. mentioned in commit messages) that confidence has rubbed off on them. Given that, why would I believe it when asking something medical, legal, historical, or otherwise outside of my domain or that I can't somehow verify?

You should be able to verify facts outside of your domain of expertise. Every time a new study comes around saying “actually, everyone should be drinking 3 glasses of wine a night”, do you take a trip to the liquor store?

LLMs are far from the only thing you rely on that will confidently lie to you.

Re: Ask HN: Is anyone else bearish on OpenAI?

#190

Post by pradn »

kubiton wrote: post #184

LLM are the first interface I as.a.software engineer is impressed by. My company also got it immediately and is rolling it out globally. GitHub copilot is already helpful. GitHub copilot for docs (announced at GitHub next) is a game changer. I used openai to reformulate emails and suddenly got positive feedback about my announcement emails. I communicate with openai German and English how ever I see fit. It's very ha…

> I communicate with openai German and English how ever I see fit.

I did not foresee that you could mix input languages. That’s fascinating. Multilingual people use languages for different purposes, often in the same sentence. Ie: technical jargon in English, a joke in Arabic, etc.

I expect the difference in connotation/feeling/mood to be less relevant for an LLM, if you’re working with facts. But there was a recent post showing LLMs performing better when you said you were stressed/scared. Did you notice any such differences for your multi-lingual inputs?

Post reply on HN