3 ms·

>A shrewd plan considering he's not offering something beyond what another company with a large nvidia cluster could offer. It's been 4 months, no one has rele

>A shrewd plan considering he's not offering something beyond what another company with a large nvidia cluster could offer. It's been 4 months, no one has released anything nearly as good as the initial release of ChatGPT. Meanwhile OpenAI has released GPT-4 and is trialing plugins and 32k context. Either their competition is incompetent or OpenAI is doing something right.

endisneigh 4y ago
Without knowing the profit margin of OpenAI it’s anyone’s guess. Personally I’d be surprised if it’s profitable, azure subsidies aside
- sebzim4500 4y ago
  Who said anything about making a profit? If Google could make Bard work as well as GPT-4 they would do it, even if it had to operate at a significant loss to start.
  - endisneigh 4y ago
    If it doesn’t make a profit there’s no incentive to do it.
    - mimd 4y ago
      It's a stunt now for both companies as it's a proxy for both search, AI capabilities for sale, and clouds. But eventually they will reach a limit of payoff. But 4 months is short for product development, especially as it sounds like google was caught off guard by the public interest in interacting with chatgpt.
- polycaster 4y ago
  According to Reuters [1] in 2022 they projected $200 million revenue in 2023 and $1 billion revenue in 2024. In 2019 they Planned to spend the then 1 bn funding within the next 5 years or so. So I suppose they’re not far away from net zero at least. [1]: https://www.reuters.com/business/chatgpt-owner-openai-projects-1-billion-revenue-by-2024-sources-2022-12-15/ https://www.reuters.com/business/chatgpt-owner-openai-projec...
  - polycaster 4y ago
    Also, recently sama indicated chatgpt would make a little money (since the context of this statement is how the hell it could be profitable I assume he talked about profit). https://news.ycombinator.com/item?id=34992606 https://news.ycombinator.com/item?id=34992606
    - mimd 4y ago
      It will be interesting to see how their sales progress after the initial hype period. Some of their contracts to incorporate gpt into products sounded like initial marketing and hype, but perhaps even independent of those, they will make it back with services to bing. But the subdued reaction to gpt4's increase in capabilities means their might be a lack of incentive to pay for improved versions.
mimd 4y ago
DeepMind could no doubt argue the same for openAI failing to play go and to not pursuing protein folding. Their several years behind on that one. :) Anyway, there have been several open source versions in the gpt3-3.5 range. Those are your base to build off of. Their's just little incentive to try to go participate in a moneysink competition with Google and Microsoft.
- starik36 4y ago
  Not really. I tried several fine tuned iterations of LLaMa and none of them are even close to ChatGPT.
stoniejohnson 4y ago
What about Claude from Anthropic?
- HarHarVeryFunny 4y ago
  They're doing very well too, but the company was founded by people from OpenAI, so it'd seem they took a lot of know-how with them.
  - jimsimmons 4y ago
    Or brought that know how into OpenAI in the first place?
    - HarHarVeryFunny 4y ago
      It seems OpenAI has achieved more with this tech than anyone else, and this level of capability didn't exist before OpenAI, so there's a limit to what can have come from outside. It seems that Ilya Sutskever is perhaps the driving force behind this, and he was certainly there from the start and presumably had a huge part in pushing this line of Transformer research before they had enough success with it to attract a lot of outside talent (OpenAI is now ~400 employees).
HarHarVeryFunny 4y ago
> Either their competition is incompetent or OpenAI is doing something right. Yeah - I'd say a bit of both. Google certainly seem to have dropped the ball, and now seem to be desperately trying to play catch up. OpenAI certainly seem to be doing a lot right though, in addition to being all-in and laser focused on this tech. It seems there's a lot more to improving these models than just scaling them up. Altman said "there's a lot of understanding went into building GPT-4", and they've invested a lot of effort into the "alignment" aspect - how to control these models to modify behavior - which is what makes the difference between a neat tech demo and something actually useful.
outside1234 4y ago
Google is going to bring something out - it might take 9 months - but they will blow the hinges off when they get there
- transcriptase 4y ago
  To what end? Other than showing the world they’re still capable of doing engineering beyond A/B testing methods of leveraging personal information to improve CTR on an increasing volume of ads they shove into products. Google home was the perfect chance to showcase their engineering/AI ability, and yet some days it still struggles to interpret a voice command its done 100 times prior, even when you verify that it heard the exact same command you used yesterday.

endisneigh 4y ago

Without knowing the profit margin of OpenAI it’s anyone’s guess. Personally I’d be surprised if it’s profitable, azure subsidies aside

sebzim4500 4y ago
Who said anything about making a profit? If Google could make Bard work as well as GPT-4 they would do it, even if it had to operate at a significant loss to start.
- endisneigh 4y ago
  If it doesn’t make a profit there’s no incentive to do it.
  - mimd 4y ago
    It's a stunt now for both companies as it's a proxy for both search, AI capabilities for sale, and clouds. But eventually they will reach a limit of payoff. But 4 months is short for product development, especially as it sounds like google was caught off guard by the public interest in interacting with chatgpt.
polycaster 4y ago
According to Reuters [1] in 2022 they projected $200 million revenue in 2023 and $1 billion revenue in 2024. In 2019 they Planned to spend the then 1 bn funding within the next 5 years or so. So I suppose they’re not far away from net zero at least. [1]: https://www.reuters.com/business/chatgpt-owner-openai-projects-1-billion-revenue-by-2024-sources-2022-12-15/ https://www.reuters.com/business/chatgpt-owner-openai-projec...
- polycaster 4y ago
  Also, recently sama indicated chatgpt would make a little money (since the context of this statement is how the hell it could be profitable I assume he talked about profit). https://news.ycombinator.com/item?id=34992606 https://news.ycombinator.com/item?id=34992606
  - mimd 4y ago
    It will be interesting to see how their sales progress after the initial hype period. Some of their contracts to incorporate gpt into products sounded like initial marketing and hype, but perhaps even independent of those, they will make it back with services to bing. But the subdued reaction to gpt4's increase in capabilities means their might be a lack of incentive to pay for improved versions.

mimd 4y ago

DeepMind could no doubt argue the same for openAI failing to play go and to not pursuing protein folding. Their several years behind on that one. :) Anyway, there have been several open source versions in the gpt3-3.5 range. Those are your base to build off of. Their's just little incentive to try to go participate in a moneysink competition with Google and Microsoft.

starik36 4y ago
Not really. I tried several fine tuned iterations of LLaMa and none of them are even close to ChatGPT.

stoniejohnson 4y ago

What about Claude from Anthropic?

HarHarVeryFunny 4y ago
They're doing very well too, but the company was founded by people from OpenAI, so it'd seem they took a lot of know-how with them.
- jimsimmons 4y ago
  Or brought that know how into OpenAI in the first place?
  - HarHarVeryFunny 4y ago
    It seems OpenAI has achieved more with this tech than anyone else, and this level of capability didn't exist before OpenAI, so there's a limit to what can have come from outside. It seems that Ilya Sutskever is perhaps the driving force behind this, and he was certainly there from the start and presumably had a huge part in pushing this line of Transformer research before they had enough success with it to attract a lot of outside talent (OpenAI is now ~400 employees).

HarHarVeryFunny 4y ago

> Either their competition is incompetent or OpenAI is doing something right. Yeah - I'd say a bit of both. Google certainly seem to have dropped the ball, and now seem to be desperately trying to play catch up. OpenAI certainly seem to be doing a lot right though, in addition to being all-in and laser focused on this tech. It seems there's a lot more to improving these models than just scaling them up. Altman said "there's a lot of understanding went into building GPT-4", and they've invested a lot of effort into the "alignment" aspect - how to control these models to modify behavior - which is what makes the difference between a neat tech demo and something actually useful.

outside1234 4y ago

Google is going to bring something out - it might take 9 months - but they will blow the hinges off when they get there

transcriptase 4y ago
To what end? Other than showing the world they’re still capable of doing engineering beyond A/B testing methods of leveraging personal information to improve CTR on an increasing volume of ads they shove into products. Google home was the perfect chance to showcase their engineering/AI ability, and yet some days it still struggles to interpret a voice command its done 100 times prior, even when you verify that it heard the exact same command you used yesterday.