One obvious business case is customer service chatbots, which all of a sudden do seem to have got loads better, ie, they work for the first time and can at least interact and understand what your problem is (before passing you on to a human) which has been 30 years coming.
i agree that's a killer business solution for customer servicing
but it's limited as i said, let's say amazon can replace 90% labor of the customer service department with chatbots, it's still insignificant compared with the ai vision described by the industry
You would probably say the same thing about Amazon when it was founded in 1994. Personally, I don’t agree with the article if for no other reason than we’re too early in the cycle to know the outcomes. I also don’t think cheaper models will sink the AI market because really models are just infrastructure. It’s what you build on top of the models that’s useful. For example, ChatGPT is actually an agent built on top of models.
> we’re too early in the cycle to know the outcomes
who knows, maybe 10years later, the weekly work hour will dropped to 20 from 40, so that consumers can give more their time and money-per-workhour to corporations, or workers must buy more equipments to earn a life
SEO, Marketing, Content Generation(Text,Images,Videos) has completely changed by AI. Lot of new companies are getting created in this space. Automation via computer use is the new set of application I am seeing in the market.
It is very easy now to analyze what is happening in a computer screen and make rule based application on what to do next. Lot of jobs will be disrupted by it.
SEO, marketing, and content generation are fundamentally parasitic bullshit. Unlike the initial .com wave, there is value to some companies, but basically no value to end consumers or (ultimately) humanity.
I do consulting for multiple GPU hosting companies. DeepSeek means more things can be done with less and the demand for access to GPU has only increased.
The analysts, analysts who do not code, analysts who don't run GPU clouds, are wrong.
This is my intuition as well. DeepSeek and the proliferation of local LLMs have showed us just how much can be done with relatively light hardware. It finally has me wanting to build a rig for local use when before I felt like the technology wasn't quite there for me to do anything too useful with some consumer hardware. This changed over night. I imagine there are many folks like me.
There's also lot you can do with not-brand-new server hardware that doesn't cost that much. The hard part is a location to put it where the noise is not bothersome. Though many who are "upper middle class" or on low-cost private land can manage to make this happen. I have seen people that don't have a lot of money, rather young people, do a lot on private land. Small dedicated server buildings are a thing.
Open source and/or low precision LLMs have been around for years and people have been hacking on them and doing all sorts of things with them. AI is incredibly overhyped and was due for commoditization. If a Chinese company releasing yet another decent but somehow this time incredibly overhyped LLM is what it takes to get investors to see the writing on the wall, then good riddance.
It would be well-deserved and helpful towards culling many AI services, which are essentially snake oil. For example, the science and research on lie detection using image and audio analysis is tenuous at best, but there companies purporting to offer services such as these. On the other hand, such services are perhaps useful for milking money and resources from unscrupulous actors whose primary goal is “cover your ass”. So, essentially the McKinsey business model.
On the contrary with cheap LLMs this practice will be more prevalent. If you lift the cost barrier the McKinseys will be attaching all sorts of AI to everything.
All the AI bust means is that for engineers working at companies that invested too far into AI they're going to be looking for jobs. ML departments will be culled too. Just like the dot-com bust.
For anyone who doesn't realise how similar this is to the dotcom bubble, here is a thought experiment.
In 1998, you would have spent 2 years hearing about Altavista and the new search engine paradigm with curated hyperlinks. They are buying lots and lots of Cisco gear to serve their network. You get excited and invest $100B into Altavista and Cisco.
Then all of a sudden, a startup called Google shows up with a cheaper and more efficient way of presenting search, with much less networking equipment.
What do you think happens to the $100B invested with Altavista and Cisco? What do you think happens to OpenAI and Nvidia?
I just made a very similar analogy to someone yesterday (we work at Google).
Open source stacks and commodity hardware win. What do you think the next generation of engineers and scientists is training on right now, as broke students? They aren’t paying licenses and big API credits.
Yeah, but OpenAI still has massive amounts of hardware. Maybe their models are obsolete now, but if they use Deepseek's model, they might be able to make it much better much more cheaply than before.
I think a big difference between dotcom bubble and ai bubble is:
where is the business/how it profit?
there're solid new-born business with dotcom, like amazon and microsoft
but who are the customers of AI(or LLM, to be specific)? for now i can only see limited user such as programers
how will business gain profits by LLMs
and it feels like users are already consumed totally by existing internet and apps, it's a stock market rather than an incremental market
This is not like a gold mine and shovels, it's more like a cliff and ropes
One obvious business case is customer service chatbots, which all of a sudden do seem to have got loads better, ie, they work for the first time and can at least interact and understand what your problem is (before passing you on to a human) which has been 30 years coming.
i agree that's a killer business solution for customer servicing
but it's limited as i said, let's say amazon can replace 90% labor of the customer service department with chatbots, it's still insignificant compared with the ai vision described by the industry
You would probably say the same thing about Amazon when it was founded in 1994. Personally, I don’t agree with the article if for no other reason than we’re too early in the cycle to know the outcomes. I also don’t think cheaper models will sink the AI market because really models are just infrastructure. It’s what you build on top of the models that’s useful. For example, ChatGPT is actually an agent built on top of models.
chatgpt doesn't bring profits to openai/microsoft
but i agree:
> we’re too early in the cycle to know the outcomes
who knows, maybe 10years later, the weekly work hour will dropped to 20 from 40, so that consumers can give more their time and money-per-workhour to corporations, or workers must buy more equipments to earn a life
SEO, Marketing, Content Generation(Text,Images,Videos) has completely changed by AI. Lot of new companies are getting created in this space. Automation via computer use is the new set of application I am seeing in the market.
It is very easy now to analyze what is happening in a computer screen and make rule based application on what to do next. Lot of jobs will be disrupted by it.
the problem is, how much users are willing to pay for an AI-powered google or an AI-powered office365
and searching is not google's business, advertising is
the most profitable business of AI seems always be 2B but not 2C, means it's hard to grab more value from consumption of customers
SEO, marketing, and content generation are fundamentally parasitic bullshit. Unlike the initial .com wave, there is value to some companies, but basically no value to end consumers or (ultimately) humanity.
This was exactly the same in the dotcom bubble. Like exactly the same.
I do consulting for multiple GPU hosting companies. DeepSeek means more things can be done with less and the demand for access to GPU has only increased.
The analysts, analysts who do not code, analysts who don't run GPU clouds, are wrong.
The anals are wrong.
This is my intuition as well. DeepSeek and the proliferation of local LLMs have showed us just how much can be done with relatively light hardware. It finally has me wanting to build a rig for local use when before I felt like the technology wasn't quite there for me to do anything too useful with some consumer hardware. This changed over night. I imagine there are many folks like me.
There's also lot you can do with not-brand-new server hardware that doesn't cost that much. The hard part is a location to put it where the noise is not bothersome. Though many who are "upper middle class" or on low-cost private land can manage to make this happen. I have seen people that don't have a lot of money, rather young people, do a lot on private land. Small dedicated server buildings are a thing.
Or, we just end up in a Kowloon Walled City cyberpunk hellscape with water cooling pipes and fans running though all our environment.
This may not affect the total demand, but the distribution of demand. What's also affected, in my opinion:
- individual company valuations
- capacity planning models ("what would it take to do x?") and related funding
- overall competitive landscape (yes you can compete with OpenAI after all without a trillion dollars in funding)
- it's not the coveted high-end GPUs anymore, more affordable/available hardware may also work
Open source and/or low precision LLMs have been around for years and people have been hacking on them and doing all sorts of things with them. AI is incredibly overhyped and was due for commoditization. If a Chinese company releasing yet another decent but somehow this time incredibly overhyped LLM is what it takes to get investors to see the writing on the wall, then good riddance.
It would be well-deserved and helpful towards culling many AI services, which are essentially snake oil. For example, the science and research on lie detection using image and audio analysis is tenuous at best, but there companies purporting to offer services such as these. On the other hand, such services are perhaps useful for milking money and resources from unscrupulous actors whose primary goal is “cover your ass”. So, essentially the McKinsey business model.
On the contrary with cheap LLMs this practice will be more prevalent. If you lift the cost barrier the McKinseys will be attaching all sorts of AI to everything.
All the AI bust means is that for engineers working at companies that invested too far into AI they're going to be looking for jobs. ML departments will be culled too. Just like the dot-com bust.
The fact that o3 exists moots any point that relies on AI becoming cheap. When/if deepseek
At most this will force OpenAI to hire some real engineers and not just new grads who are easy to exploit.
OpenAI pays close to a million per year. Anybody can be exploited at that salary.
“Pays”
Are those PPUs vesting and is there a cash market for them?
For anyone who doesn't realise how similar this is to the dotcom bubble, here is a thought experiment.
In 1998, you would have spent 2 years hearing about Altavista and the new search engine paradigm with curated hyperlinks. They are buying lots and lots of Cisco gear to serve their network. You get excited and invest $100B into Altavista and Cisco.
Then all of a sudden, a startup called Google shows up with a cheaper and more efficient way of presenting search, with much less networking equipment.
What do you think happens to the $100B invested with Altavista and Cisco? What do you think happens to OpenAI and Nvidia?
I just made a very similar analogy to someone yesterday (we work at Google).
Open source stacks and commodity hardware win. What do you think the next generation of engineers and scientists is training on right now, as broke students? They aren’t paying licenses and big API credits.
Yeah, but OpenAI still has massive amounts of hardware. Maybe their models are obsolete now, but if they use Deepseek's model, they might be able to make it much better much more cheaply than before.