I don't know about others, but I have learned at this point that relying on proprietary walled-garden software is a terrible idea.
Ask HN: Is anyone else bearish on OpenAI?
291–300 of 321 posts
Re: Ask HN: Is anyone else bearish on OpenAI?
#292Earlier quoted context omitted.
Please, please, reference the tasks or chats that were especially impressive. For example, trying to get chat gpt to do something very simple for my work, like implementing a convolution it took me in circles and circles. It gets the general idea right, sure. But it actually makes significant minor errors that ended up being more confusing than helpful.
I asked GPT-4 to implement a convolution layer in Numpy, including backpropagation. It produced correct, working code within ~15 minutes of prompting. Are you sure you used GPT-4?
Here's an example recreation.
https://chat.openai.com/share/b80aa985-0d88-4cf3-8c87-c2047f...
I admittedly give it a vague question "where the signals overlap"
In context of the original thing I was trying to do, I was filtering a receive signal and trying to remove the delay from the convolution
Re: Ask HN: Is anyone else bearish on OpenAI?
#293Earlier quoted context omitted.
We need to stop propagating this nonsense. GPT-4 still messes up.
It's not nonsense at all. The degree to which GPT-4 messes up is dramatically lower than 3.5.
The nonsense is the automatic conclusion that when ChatGPT messes up, it must be 3.x; which is demonstrably false.
Re: Ask HN: Is anyone else bearish on OpenAI?
#294> This feels a lot like crypto Why do I keep hearing this and where is it coming from? I hear it so often that it feels like an agenda being pushed but I can't imagine from whom.
I think it's just that crypto was the most recent hyped thing. To me the developments in AI feel more like 1995 for the web. It was just starting to get useful and was obviously a thing with a big future but a lot of the early companies were flakey.
Nowadays, all the major browser engines are open-source. Hopefully that will also become the case with AI.
Re: Ask HN: Is anyone else bearish on OpenAI?
#295Earlier quoted context omitted.
Are you using GPT-3.5 or GPT-4? A hallucination like FableGL sounds like 3.5, which is not something you should base your expectations of the tech on. Also, although I haven't seen it before, the Warren Abstract Machine seems like too big of a job for GPT-4. It excels at smaller tasks such as "Convert this C code to idiomatic Rust", or "Write a SQL query that grabs records from X, joins with Y, filters by Z". You mig…
> A hallucination like FableGL sounds like 3.5, which is not something you should base your expectations of the tech on. So in order to properly evaluate a service, we need to already be subscribed to it in the first place?
Yes. Or just use the api and pay for usage (but in that case It could also get more complicated since you have full control)
Re: Ask HN: Is anyone else bearish on OpenAI?
#296I'm a grad student and I don't really use it for my job but ChatGPT is something millions of people use daily so I think it has a future.
If it becomes as big as Google depends on what Open AI does, Google didn't become as massive as it is just off search.
Re: Ask HN: Is anyone else bearish on OpenAI?
#297Earlier quoted context omitted.
> I think this speaks to the bubble you're in and how your kind of software development probably represents 0.1% of what's out there. Grandparent specifically said they find ChatGPT useful for things outside their wheelhouse: >> [...]troubleshooting situations that flow into me - working in areas that I have the slimmest of backgrounds in It sounds to me like it may not be safe to use ChatGPT if you can't check its w…
Not the original poster, but here are some examples: - Paste a bunch of log code about which I have no idea. Ask to identify and explain the problem. - Wireshark / dmesg / OpenWrt configuration pasting, ask to fix the problem. For instance, I fixed a Wi-Fi issue in a heterogeneous setup, which turned out to be caused by a stray DHCPv6 server. - Paste C code, along with an error log, and ask to fix the problem. - Past…
Can you paste a transcript for any of those examples here? I’m genuinely curious. I’ve never seen a single actual transcript from anyone talking about how they use ChatGPT all the time for programming work and it’s great.
Re: Ask HN: Is anyone else bearish on OpenAI?
#298Earlier quoted context omitted.
Not the original poster, but here are some examples: - Paste a bunch of log code about which I have no idea. Ask to identify and explain the problem. - Wireshark / dmesg / OpenWrt configuration pasting, ask to fix the problem. For instance, I fixed a Wi-Fi issue in a heterogeneous setup, which turned out to be caused by a stray DHCPv6 server. - Paste C code, along with an error log, and ask to fix the problem. - Past…
Yeah I think the people who haven't found value are people who either don't have strong communication skills hence cannot articulate what they need OR work in very speciailized fields - either way they're clearly the minority seeing how many paying users chatgtp has and how fast they got there
> Yeah I think the people who haven't found value are people who either don't have strong communication skills hence cannot articulate what they need OR work in very speciailized fields
This is so weird and seems to be a common thread. You, and others, seem to be saying that ChatGPT is not useful if you're too dumb or too smart. It just seems both tautological and strange.
My communication skills are fine. ChatGPT is just not useful for me, period. I've tried taking what people have mentioned here, trying to have a so-called dialogue with it when things it suggests, for much more simple things, aren't quite correct. It keeps suggesting all these things, but it just feels like an interactive StackOverflow, except all the suggestions are wrong or don't work.
Re: Ask HN: Is anyone else bearish on OpenAI?
#299Earlier quoted context omitted.
Yeah I think the people who haven't found value are people who either don't have strong communication skills hence cannot articulate what they need OR work in very speciailized fields - either way they're clearly the minority seeing how many paying users chatgtp has and how fast they got there
> don't have strong communication skills hence cannot articulate Basically this - even in a highly specialised field you would still find value in it if you're providing enough context and describing the problem properly. People don't want to hear that they're poor communicators - it's much easier to blame the tech.
> People don't want to hear that they're poor communicators - it's much easier to blame the tech.
It's just hard to even know what to say to this. So you are implying that the problem lies with the human, who holds years of experience in education, problem solving, communication, etc., and not the "AI", which has no experience in any of that and has no real thought processes or problem solving or reasoning abilities? And people say ChatGPT is over-hyped!
Re: Ask HN: Is anyone else bearish on OpenAI?
#300Earlier quoted context omitted.
It's not nonsense at all. The degree to which GPT-4 messes up is dramatically lower than 3.5.
The nonsense is the automatic conclusion that when ChatGPT messes up, it must be 3.x; which is demonstrably false.
But no one said that. GP just said they were different products. And that is true. It is meaningless to show an error that GPT 3.5 produced in a conversation about 4. They are separate products.