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

Certified 100% AI-free organic content

substack.piszek.com

Post reply on HN
Print view View original

161–170 of 275 posts

Re: Certified 100% AI-free organic content

#161

Post by spacephysics »

sillysaurusx wrote: post #114
eloisius wrote: post #112

Earlier quoted context omitted.

> Could we end up having AI quality trend downwards due to AI ingesting its own old outputs and reinforcing bad habits? I think it's a particular risk for text generation. This is exactly what I don't like about Copilot, maybe even more than the IP ethics of it. If it really succeeds, it's going to have a feedback loop that amplifies its own code suggestions. The same boilerplate-ish kind of code that developers gene…

Devs can use their noggins to check whether copilot’s output is decent or not. It’s not a given that people will use it automatically.

I think many developers of today agree.

Todays programmers can see copilot output and probably think “well that’s not optimal”. Fast forward five years, new CS grads are using Copilot 3.0, and are used to specific auto-completes that copilot gives for certain tasks, as they may have never needed to go beyond some of the more basic suggestions.

It “feels” like an older programmer seeing a younger web dev and going “you’re wasting MB of memory!”

While true the web has gotten slower in many regards, and indeed memory may have been wasted, business value creation typically doesn’t care if a few MB is sub optimally wasted, while the previous generation does.

Re: Certified 100% AI-free organic content

#162

Post by spacephysics »

boringg wrote: post #130
sillysaurusx wrote: post #114

Earlier quoted context omitted.

Devs can use their noggins to check whether copilot’s output is decent or not. It’s not a given that people will use it automatically.

That and I bet the generic boilerplate code that co-pilot produces is on average much better than the boiler plate code that the average dev might use. So it removes the lower level work.

I think that can surely be the case, but just like any AI there may need to be manual review to asses what is the most optimal way to go about task X, and retrain.

I can see this go both ways:

boilerplate code being a great generic solution to a set of problems, but a more seasoned programmer may say “that works, but for our use case the trade offs don’t make sense”

Or alternatively, “this code wasn’t something I knew I could do in language X, and it’s far more efficient”

Re: Certified 100% AI-free organic content

#163

Post by rcme »

CuriouslyC wrote: post #144
rcme wrote: post #126

Earlier quoted context omitted.

Current LLMs are trained on as much data that can be scraped from the public internet. It’s simply not possible to annotate that much data, even with crowdsourcing. It’s not even a matter of cost. You’d basically need to duplicate the amount of data on the internet. I don’t think you’re appreciating the scale of the data involved in training these models.

Not necessarily. The bloom model (a GPT competitor and similarly sized) was trained on 1.5T of text, which reduces down to 350B unique tokens. If you took a histogram of those unique tokens, it would have a very long tail with probably 1% or less being well represented. That leaves 350M common tokens to serve as the basis for token tuples being fed into crowdsourcing. There are probably ~2-5B very common token sequen…

BLOOM isn’t as good as GPT-3 because it doesn’t use as much training data. LLM quality is still data bound [0]. Further limiting data by requiring annotation is not going to work, at least with the current LLM modeling approach.

0: https://www.alignmentforum.org/posts/6Fpvch8RR29qLEWNH/chinc...

Re: Certified 100% AI-free organic content

#164

Post by layer8 »

alpos wrote: post #133
gattilorenz wrote: post #84

Earlier quoted context omitted.

It’s as if ChatGPT was behaving like a language model with no real connection or understanding of R… hmmmmm…

This take strikes me as a little off. Programming languages are language. Unlike natural languages they are also based on context-free grammar. So an understanding of programming languages should actually be easier for even a general language model to incorporate than natural languages. We can expect a bot like this to not really get context clues in natural language, although they seem to be getting better at that,…

The failure to realize that the API doesn’t exist and therefore the code won’t work in practice, however, is a major lack of insight and understanding.

Re: Certified 100% AI-free organic content

#165

Post by lolinder »

luxcem wrote: post #152
dale_glass wrote: post #15

> Published content will be later used to train subsequent models, and being able to distinguish AI from human input may be very valuable going forward I find this to be a particularly interesting problem in this whole debacle. Could we end up having AI quality trend downwards due to AI ingesting its own old outputs and reinforcing bad habits? I think it's a particular risk for text generation. I've already run into…

> Published content will be later used to train subsequent models, and being able to distinguish AI from human input may be very valuable going forward Every discussion on AI take the example of ChatGPT and its inherent flaws but AI-generated content doesn't have to be dull and low quality. One question that bother me is does it really matter? If AI-generated content is on par with Human-made or even better does it m…

> Every discussion on AI take the example of ChatGPT and its inherent flaws but AI-generated content doesn't have to be dull and low quality.

To get away from that we'd have to dramatically change our approach. The LLMs we have are trained on as much content as possible and essentially average out the style of their training data. What it writes reads like a B-grade high school essay because that is what you get when you average all the writing on the internet.

It's not obvious to me that a creative approach that boils down to "pick the most likely next word given the context so far" can avoid sounding bland.

Re: Certified 100% AI-free organic content

#166

Post by gattilorenz »

alpos wrote: post #133
gattilorenz wrote: post #84

Earlier quoted context omitted.

It’s as if ChatGPT was behaving like a language model with no real connection or understanding of R… hmmmmm…

This take strikes me as a little off. Programming languages are language. Unlike natural languages they are also based on context-free grammar. So an understanding of programming languages should actually be easier for even a general language model to incorporate than natural languages. We can expect a bot like this to not really get context clues in natural language, although they seem to be getting better at that,…

My answer was mildly tongue in cheek, and I see where you’re going.

On the other hand, one of the other posters asked “to generate a parallax effect in Qt/QML. It simply used a QML Elemened with the name Parallax”. Is this an insight, or is this answering “yes, I could” to “could you pass me the salt?”. Maybe the line between the two is a fine one, and I didn’t realize that yet.

In general, it feels like copying part of the question (“write parallax code”) in the answer is the easy part of the task…

Re: Certified 100% AI-free organic content

#167

Post by layer8 »

luxcem wrote: post #152
dale_glass wrote: post #15

> Published content will be later used to train subsequent models, and being able to distinguish AI from human input may be very valuable going forward I find this to be a particularly interesting problem in this whole debacle. Could we end up having AI quality trend downwards due to AI ingesting its own old outputs and reinforcing bad habits? I think it's a particular risk for text generation. I've already run into…

> Published content will be later used to train subsequent models, and being able to distinguish AI from human input may be very valuable going forward Every discussion on AI take the example of ChatGPT and its inherent flaws but AI-generated content doesn't have to be dull and low quality. One question that bother me is does it really matter? If AI-generated content is on par with Human-made or even better does it m…

It’s not clear when and how we would reach that level of quality. It doesn’t seem very relevant to the present state of affairs.

Re: Certified 100% AI-free organic content

#168

Post by darkerside »

dwringer wrote: post #27
darkerside wrote: post #21

> do you really want to cry watching a movie that was 100% produced by robots? The fear reminds me of the early days of hip hop, where songs remixed from the past were blasted as unoriginal. I think we've all mostly agreed now that you can build new content that honors the old whole being completely fresh.

This sums it up perfectly in my mind, the analogy is extremely apt IMHO. As a musician I find it impossible to spontaneously produce melodies, rhythms, etc, that aren't heavily influenced by what I've listened to - perfectly analogous to how AI generated art is heavily influenced by the billions of training images it's ingested. The difficulty in composing is finding a way to transform, combine, and synthesize the in…

I've only got a single down vote. I assume someone just didn't get it, but I'm glad you did!

Re: Certified 100% AI-free organic content

#169

Post by luxcem »

thenerdhead wrote: post #134

The point about emotional response is good. I'm not sure how to best describe it, but every time I interact with AI, there is very little emotional response from it. Rather it's a "good enough" response rather than a sense of awe or horror. I've been experimenting with writing recently and wanting to provide some AI imagery to match the emotions I'm expressing. A painting like "Wanderer above the Sea of Fog" evokes m…

Do you think your emotional response is related to the external knowledge of who produced the painting or is it only based on the visual impression?

We probably can still figure out if a painting is original or AI-generated but I don't think we can much longer as AI improve.

The question would be could we feel emotions even if the source material is artificial. I think the answer is yes. Human brain is can easily be tricked.

Re: Certified 100% AI-free organic content

#170

Post by htrp »

imranq wrote: post #121

OpenAI is working on a watermark for their models (if not already there) that would recognize content as AI generated. If its good enough, it should be able to filter out AI content when training new models https://techcrunch.com/2022/12/10/openais-attempts-to-waterm...

The problem is that you can always use something else to watermark it.
Post reply on HN