5 ms·

It is weird that one of the most valued markets (openai, microsoft investments, nvidia gpus, ...) is based on a stack that is available to anyone that can pay f

It is weird that one of the most valued markets (openai, microsoft investments, nvidia gpus, ...) is based on a stack that is available to anyone that can pay for the resources to train the models and that in my opinion still has to deliver to the expectations that have been created around it. Not saying it is a bubble but something seems imbalanced here.

throwaway42668 2y ago
It's okay to say it. It's a bubble. It was just the next in line to be inflated after crypto.
- superluserdo 2y ago
  I wouldn't write it off as a bubble, since that usually implies little to no underlying worth. Even if no future technical progress is made, it has still taken a permanent and growing chunk of the use case for conventional web search, which is an $X00bn business.
  - thegeomaster 2y ago
    A bubble doesn't necessarily imply no underlying worth. The dot-com bubble hit legendary proportions, and the same underlying technology (the Internet) now underpins the whole civilization. There is clearly something there, but a bubble has inflated the expectations beyond reason, and the deflation will not be kind on any player still left playing (in the sense of AI winter), not even the actually-valuable companies that found profitable niches.
- KaiserPro 2y ago
  I mean OpenAI is a bubble, if it pops, its big enough to take the rest of tech with it.
  - throwaway42668 2y ago
    Alternatively, it might make capital available to other things. It's at least theoretically possible that all the liquidity and leverage in the top of the market could tire itself of chasing the next tulip mania. For instance, $6 Billion could have gone into climate tech instead of ElizaX. My problem with these dumb hype cycles is all the other stuff that gets starved in their wake.
jasode 2y ago
>one of the most valued markets ... is based on a stack that is available to anyone The sophisticated investors are not betting on future increasing valuations based on current LLMs or the next incremental iterations of it. That's a "static" perspective based on what outsiders currently see as a specific product or tech stack. Instead, you have to believe in a "dynamic" landscape where OpenAI the organization of employees can build future groundbreaking models that are not LLMs but other AI architectures and products entirely. The so-called "moat" in this thinking would be the "OpenAI team to keep inventing new ideas beyond LLM". The moat is not the LLM itself. Yes, if everyone focuses LLMs, it does look like Meta's free Llama models will render OpenAI worthless. (E.g. famous memo : https://www.google.com/search?q=We+have+no+Moat%2C+and+Neither+does+OpenAI https://www.google.com/search?q=We+have+no+Moat%2C+and+Neith...) As an analogy, imagine that in the 1980s, Microsoft's IPO and valuation looks irrational since "writing programming code on the Intel x86 stack" is not a big secret. That stock analysis would then logically continue saying "Anybody can write x86 software such as Lotus, Borland, etc." But the lesson learned was that the moat was never the "Intel x86 stack"; the moat was really the whole Microsoft team. That said, if OpenAI doesn't have any future amazing ideas, their valuation will crash.
- melenaboija 2y ago
  I agree in most of what you said. The main problem for me is that I dont see LLMs are as solid foundations to create a company as technology progress from the 80s. Im 42 though and already feeling old to understand the future lol
- silvestrov 2y ago
  I'd say that Microsoft's moat was the copyright law and ability to bully the hardware companies with exclusive distribution contracts. Writing a new DOS (or Windows 3) from scratch is something a lot of developers could do. They just couldn't do it legally. And thus it was easy to bully Compaq and others into only distributing PCs with DOS/Windows installed. For some time you even had to pay the Microsoft fee when you wanted a PC with Linux installed.
  - Schiendelman 2y ago

throwaway42668 2y ago

It's okay to say it. It's a bubble. It was just the next in line to be inflated after crypto.

superluserdo 2y ago
I wouldn't write it off as a bubble, since that usually implies little to no underlying worth. Even if no future technical progress is made, it has still taken a permanent and growing chunk of the use case for conventional web search, which is an $X00bn business.
- thegeomaster 2y ago
  A bubble doesn't necessarily imply no underlying worth. The dot-com bubble hit legendary proportions, and the same underlying technology (the Internet) now underpins the whole civilization. There is clearly something there, but a bubble has inflated the expectations beyond reason, and the deflation will not be kind on any player still left playing (in the sense of AI winter), not even the actually-valuable companies that found profitable niches.
KaiserPro 2y ago
I mean OpenAI is a bubble, if it pops, its big enough to take the rest of tech with it.
- throwaway42668 2y ago
  Alternatively, it might make capital available to other things. It's at least theoretically possible that all the liquidity and leverage in the top of the market could tire itself of chasing the next tulip mania. For instance, $6 Billion could have gone into climate tech instead of ElizaX. My problem with these dumb hype cycles is all the other stuff that gets starved in their wake.

jasode 2y ago

>one of the most valued markets ... is based on a stack that is available to anyone The sophisticated investors are not betting on future increasing valuations based on current LLMs or the next incremental iterations of it. That's a "static" perspective based on what outsiders currently see as a specific product or tech stack. Instead, you have to believe in a "dynamic" landscape where OpenAI the organization of employees can build future groundbreaking models that are not LLMs but other AI architectures and products entirely. The so-called "moat" in this thinking would be the "OpenAI team to keep inventing new ideas beyond LLM". The moat is not the LLM itself. Yes, if everyone focuses LLMs, it does look like Meta's free Llama models will render OpenAI worthless. (E.g. famous memo : https://www.google.com/search?q=We+have+no+Moat%2C+and+Neither+does+OpenAI https://www.google.com/search?q=We+have+no+Moat%2C+and+Neith...) As an analogy, imagine that in the 1980s, Microsoft's IPO and valuation looks irrational since "writing programming code on the Intel x86 stack" is not a big secret. That stock analysis would then logically continue saying "Anybody can write x86 software such as Lotus, Borland, etc." But the lesson learned was that the moat was never the "Intel x86 stack"; the moat was really the whole Microsoft team. That said, if OpenAI doesn't have any future amazing ideas, their valuation will crash.

melenaboija 2y ago
I agree in most of what you said. The main problem for me is that I dont see LLMs are as solid foundations to create a company as technology progress from the 80s. Im 42 though and already feeling old to understand the future lol
silvestrov 2y ago
I'd say that Microsoft's moat was the copyright law and ability to bully the hardware companies with exclusive distribution contracts. Writing a new DOS (or Windows 3) from scratch is something a lot of developers could do. They just couldn't do it legally. And thus it was easy to bully Compaq and others into only distributing PCs with DOS/Windows installed. For some time you even had to pay the Microsoft fee when you wanted a PC with Linux installed.
- Schiendelman 2y ago