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

OpenAI’s policies hinder reproducible research on language models

aisnakeoil.substack.com

Post reply on HN
Print view View original

41–50 of 394 posts

Re: OpenAI’s policies hinder reproducible research on language models

#41

Post by throwaway1851 »

atleastoptimal wrote: post #32

I understand any individual's company anti-competitive measures. OpenAI looks at Google the same way Apple looked at IBM in the 80s. What I'm worried about is a lot of the talk about guarding models, public safety and misuse of models will end up leading every big company to pull public access of their APIs. We might look at 2022-2023 as a brief golden age when regular people could use stuff like GPT-4 before it was…

A concern I have about OpenAI is that, if you're using their APIs to develop an application, they can mine your data to compete with you, or even beat you to market. They can do this indirectly, by sharing information with preferred business partners. The conflict of interest, combined with the lack of robust data privacy guarantees, makes me queasy.

If serving up generic LLM APIs becomes commoditized -- and I think it will -- they will want to monetize in other ways.

Re: OpenAI’s policies hinder reproducible research on language models

#42

Post by alwayslikethis »

saurik wrote: post #19
mvuksano wrote: post #14

I'm confused why people expect this stuff to be free? I'm surprised OpenAI was so open about their research so far. I don't blame them at all for not publishing the information. This stuff costs real money.

It might be less confusing if you consider that OpenAI was originally a non-profit. That it was even possible for them to end up in this state has massively undermined any trust I have in non-profits as a steward. https://www.vice.com/en/article/5d3naz/openai-is-now-everyth... > OpenAI was founded in 2015 as a nonprofit research organization by Altman, Elon Musk, Peter Thiel, and LinkedIn cofounder Reid Hoffman, amon…

Hi, saurik!

Yeah, I think this is a betrayal to the public. There isn't anything open about OpenAI anymore.

Re: OpenAI’s policies hinder reproducible research on language models

#43

Post by Eji1700 »

Open AI has been doing sketchyish things long before Chat GPT, and I think it's something people are eventually going to notice more and more (then again people were swearing that Musk walked on water for waaaaaay too long given his actions so fuck if I know).

They're 100% marketing FIRST. I don't think they'll outright lie, but they will absolutely screw with their data in such a way to make it look waaay more impressive than it is....which is really annoying to me because they already have impressive results. Sorta like if you managed to send a ship with people on it to mars, but kept claiming you landed on jupiter.

Re: OpenAI’s policies hinder reproducible research on language models

#44

Post by numberalltheway »

It's even more frustrating that, from what I can tell, there is nothing published about how GPT-4 improved.

I take specific exception to the hiding of the data and techniques used to generate the model. There must be something specific going on in the model that is allowing it to perform better than GPT-3 and better than what any contemporaries are able to produce. Not publishing this information hinders the further progress of the field as a whole.

Re: OpenAI’s policies hinder reproducible research on language models

#45

Post by bitL »

redox99 wrote: post #37
ShamelessC wrote: post #25

Earlier quoted context omitted.

Yannic Kilcher's opinion on this is likely correct. Similar parameter count, but trained for longer. The particulars of their instruction tuning/whatever-else-they-did are the real secret sauce.

Don't forget about a more efficient attention that let's them get 32k tokens of context.

It's still much worse than 1M context on 16GB VRAM with Reformer, but at the cost of inference speed. And you can use FlashAttention in your own models to get a more efficient/sparse attention now as well.

Re: OpenAI’s policies hinder reproducible research on language models

#46

Post by hhh »

numberalltheway wrote: post #44

It's even more frustrating that, from what I can tell, there is nothing published about how GPT-4 improved. I take specific exception to the hiding of the data and techniques used to generate the model. There must be something specific going on in the model that is allowing it to perform better than GPT-3 and better than what any contemporaries are able to produce. Not publishing this information hinders the further…

Look at the system card.

Re: OpenAI’s policies hinder reproducible research on language models

#47

Post by asdff »

throwaway1851 wrote: post #41
atleastoptimal wrote: post #32

I understand any individual's company anti-competitive measures. OpenAI looks at Google the same way Apple looked at IBM in the 80s. What I'm worried about is a lot of the talk about guarding models, public safety and misuse of models will end up leading every big company to pull public access of their APIs. We might look at 2022-2023 as a brief golden age when regular people could use stuff like GPT-4 before it was…

A concern I have about OpenAI is that, if you're using their APIs to develop an application, they can mine your data to compete with you, or even beat you to market. They can do this indirectly, by sharing information with preferred business partners. The conflict of interest, combined with the lack of robust data privacy guarantees, makes me queasy. If serving up generic LLM APIs becomes commoditized -- and I think…

Do you consent to that when you sign up for them? Its a microsoft product now and competitors to microsoft probably host their code on microsoft owned github without worry right now. Why start worrying now?

Re: OpenAI’s policies hinder reproducible research on language models

#48

Post by loveparade »

What's most surprising to me is that OpenAI really seems to believe that not publishing details will save them from competition. Everyone knows how these models work, and while I'm sure there is a bunch of "secret sauce" that OpenAI has built for training and fine-tuning, it's ridiculous to believe that the research community and competitors like Google and Facebook can't figure out the same. They just haven't really tried until recently because the capabilities and ROI of these models weren't obvious. No matter who you are, most of the smartest people work for someone else.

The only competitive advantage that OpenAI has here is a headstart of 6-12 months from all the infrastructure investment into training these kinds of models. Now that everyone wants to build competing models with the same capabilities, this advantage is going to disappear very quickly.

Re: OpenAI’s policies hinder reproducible research on language models

#50

Post by jonathankoren »

asdff wrote: post #47
throwaway1851 wrote: post #41

Earlier quoted context omitted.

A concern I have about OpenAI is that, if you're using their APIs to develop an application, they can mine your data to compete with you, or even beat you to market. They can do this indirectly, by sharing information with preferred business partners. The conflict of interest, combined with the lack of robust data privacy guarantees, makes me queasy. If serving up generic LLM APIs becomes commoditized -- and I think…

Do you consent to that when you sign up for them? Its a microsoft product now and competitors to microsoft probably host their code on microsoft owned github without worry right now. Why start worrying now?

Competitors to Microsoft buy the self hosting github option.
Post reply on HN