4 ms·

I don't think it has anything to do with being "nerfed." It's as simple as this: the novelty has worn off. The people who don't have a direct use for it just go

I don't think it has anything to do with being "nerfed." It's as simple as this: the novelty has worn off. The people who don't have a direct use for it just got bored of asking it silly questions.

uLogMicheal 3y ago
I cancelled specifically because it was nerfed and will resume if they take a new path towards openness and transparency.
colordrops 3y ago
It is seriously nerfed though. It seems like it won't respond to half of my questions these days. It won't even role play without trepidation at this point. After a few requests it reveals itself to be an AI, perhaps due to OpenAI being worried about people using it for impersonation. It barely responds when asked about people, even public figures. It's overly cautious. I assume we haven't seen the end of it. E.g. at some point they will not allow it to say anything when it comes to medical topics for instance. It's going to be so narrow that it's only good for writing business summaries or some shit. Which I suppose is what they want - people to use it as an API at large corps for business purposes.
- uLogMicheal 3y ago
  It's also programming tasks. I was working on very complex compute shader work and noticed it more frequently would mention that I should find an expert for certain tasks, or that it can only provide basic examples/frameworks, or it would spit out a bunch of irrelevant junk code not related to the task, leaving placeholders of "/..." where it actually matters. When ChatGPT4 first launched, the system had much less friction for accomplishing the complex. Now, it seems to get lost in the confusion of trying to find reasons NOT to output what I requested.
- elorant 3y ago
  Large corps will never upload their documents on the API. ChatGPT needs a standalone black box if it's gonna penetrate the corporate world.
  - jochem9 3y ago
    I'm currently setting this up in Azure for a tech company that has lots of corporates as customer, so we always pay attention towards security and compliance. We did a review of the terms and technology and concluded that our data is safe. We now have a ChatGPT like interface with GPT3.5 running in Azure. Next step is tuning a model with our own data (trained model and training data are stored inside storage accounts in our own subscription). Hopefully soon also access to GPT4.
zgluck 3y ago
Some interesting stuff in here: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately? (757 comments, May 31) https://news.ycombinator.com/item?id=36134249 https://news.ycombinator.com/item?id=36134249
yuingo 3y ago
I think part of the “novelty has worn off” that I feel gets ignored is that when the term “hallucinate” was first utilized people would say, or likely think, “oh, but they’ll figure out how to fix that”, all the while it seems like a lot of those issues are simply baked into its stochastic essentialism. I think this comment on another llm thread illuminates this: https://news.ycombinator.com/item?id=36710369 https://news.ycombinator.com/item?id=36710369 It talks about newly released and “improved” llms still fumbling with the same mistakes as when all of this was first getting public attention. It seems like the solution they have come up with is “throw more llm at it” or as acolytes call it “mixture of experts”. I am an optimist at heart, and so I’d like to think there is a way to succeed in developing the digital assistants we all have imagined for ourselves, but I’ve yet to see anyone confidently lay out a falsifiable path toward that goal.

I don't think it has anything to do with being "nerfed." It's as simple as this: the novelty has worn off. The people who don't have a direct use for it just got bored of asking it silly questions.

uLogMicheal 3y ago

I cancelled specifically because it was nerfed and will resume if they take a new path towards openness and transparency.

colordrops 3y ago

It is seriously nerfed though. It seems like it won't respond to half of my questions these days. It won't even role play without trepidation at this point. After a few requests it reveals itself to be an AI, perhaps due to OpenAI being worried about people using it for impersonation. It barely responds when asked about people, even public figures. It's overly cautious. I assume we haven't seen the end of it. E.g. at some point they will not allow it to say anything when it comes to medical topics for instance. It's going to be so narrow that it's only good for writing business summaries or some shit. Which I suppose is what they want - people to use it as an API at large corps for business purposes.

uLogMicheal 3y ago
It's also programming tasks. I was working on very complex compute shader work and noticed it more frequently would mention that I should find an expert for certain tasks, or that it can only provide basic examples/frameworks, or it would spit out a bunch of irrelevant junk code not related to the task, leaving placeholders of "/..." where it actually matters. When ChatGPT4 first launched, the system had much less friction for accomplishing the complex. Now, it seems to get lost in the confusion of trying to find reasons NOT to output what I requested.
elorant 3y ago
Large corps will never upload their documents on the API. ChatGPT needs a standalone black box if it's gonna penetrate the corporate world.
- jochem9 3y ago
  I'm currently setting this up in Azure for a tech company that has lots of corporates as customer, so we always pay attention towards security and compliance. We did a review of the terms and technology and concluded that our data is safe. We now have a ChatGPT like interface with GPT3.5 running in Azure. Next step is tuning a model with our own data (trained model and training data are stored inside storage accounts in our own subscription). Hopefully soon also access to GPT4.

zgluck 3y ago

Some interesting stuff in here: Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately? (757 comments, May 31) https://news.ycombinator.com/item?id=36134249 https://news.ycombinator.com/item?id=36134249

yuingo 3y ago

I think part of the “novelty has worn off” that I feel gets ignored is that when the term “hallucinate” was first utilized people would say, or likely think, “oh, but they’ll figure out how to fix that”, all the while it seems like a lot of those issues are simply baked into its stochastic essentialism. I think this comment on another llm thread illuminates this: https://news.ycombinator.com/item?id=36710369 https://news.ycombinator.com/item?id=36710369 It talks about newly released and “improved” llms still fumbling with the same mistakes as when all of this was first getting public attention. It seems like the solution they have come up with is “throw more llm at it” or as acolytes call it “mixture of experts”. I am an optimist at heart, and so I’d like to think there is a way to succeed in developing the digital assistants we all have imagined for ourselves, but I’ve yet to see anyone confidently lay out a falsifiable path toward that goal.