This is based on HSBC's model, which assumes some incredible numbers, such as:
> user numbers on an S-curve that by 2030 reaches 3bn, “equivalent to 44 per cent of the world’s adult population” ex China.
Unfounded statements (outside of language tasks, fwiw), such as:
>LLM subscriptions will become “as ubiquitous and useful as Microsoft 365”, HSBC says.
As well as this bold claim about OAI's potential to double the conversion rate:
>It models that by 2030, 10 per cent of OpenAI users will be paying customers, versus an estimated 5 per cent currently.
Does not include a major player in its market share analysis at all:
>Google is excluded entirely
And, still, it suggests that:
> OpenAI is expected to still be subsidising its users well into next decade
The numbers in the article are huge. It really shows how expensive this stage of AI development has become. I’m curious how long companies can keep operating at that level of burn.
The tech giants have a lot of cash-flow to burn, and at this point it's better for their investors that they do rather than admit it was a loss or the stock value would fall.
It might be a huge pile of cards but this time they have the coffers for it, if/when it falls it will be from one of the fake it till you make it companies like openai that are built on promises (and frankly their revenue numbers are no where near their promised spend so they seem dependant on more and more deals)
The team also assumes LLM companies will capture 2 per cent of the digital advertising market in revenue, from slightly more than zero currently.
This seems quite low. Meta has 3.5 billion users. ChatGPT is at 1 billion so far. By 2030, let's just stay ChatGPT reaches 2 billion years or 57% of Meta's current users.
I'd like to think that OpenAI's digital ad revenue should reach 10% by 2030 an then accelerate from there. In my opinion, the data that ChatGPT has on a user is better than the inferred user data from Instagram/FB usage. I think ChatGPT can build a better advertisement profile of each user than Meta can which can lead to better ad targeting.
This doesn't account for OpenAI's other ambitions such as Sora app.
Hey Sam Altman or OpenAI employee, if you are reading this, I think you should buy the North American version of TikTok if the opportunity presents itself. The future of short videos will be heavily AI generated/assisted. Combine Tiktok's audience with your Sora tools and ChatGPT data and you got yourself a true Instagram competitor immediately. If the $14b sales price of US Tiktok is real, that's an absolute bargain in the grand scheme of things.
From what I've seen, tons of garbage floats to the top of average non technical users. I'm not sure that Sora slop would be much worse than Daily Mail hate news slop. Or jihadist preacher slop. Or russbot misinformation slop. There's even tons of AI voiceover slop already. Sora will just be a small step towards the dead internet.
this comment feels so eerie as I am currently reading Zuboff's "The Age of Surveillance Capitalism," which itself is interesting to read now since its written before the huge AI leap.
Also, it reminded me of the following quote, mentioned in the book, from Langdon Winner
The changes and disruptions that an evolving technology repeatedly caused in modern life were accepted as given or inevitable simply because no one bothered to ask whether there were other possibilities.
This is based on HSBC's model, which assumes some incredible numbers, such as: > user numbers on an S-curve that by 2030 reaches 3bn, “equivalent to 44 per cent of the world’s adult population” ex China.
Unfounded statements (outside of language tasks, fwiw), such as: >LLM subscriptions will become “as ubiquitous and useful as Microsoft 365”, HSBC says.
As well as this bold claim about OAI's potential to double the conversion rate: >It models that by 2030, 10 per cent of OpenAI users will be paying customers, versus an estimated 5 per cent currently.
Does not include a major player in its market share analysis at all: >Google is excluded entirely
And, still, it suggests that: > OpenAI is expected to still be subsidising its users well into next decade
Fascinating.
Still, it's only 40bn per year divided by 3b so equal to around 15$/person/year
Isn't that super cheap? Just think of the revolutionary impact it would have on education, health, work etc.
I don't understand how anyone can call it a bubble.
> I don't understand how anyone can call it a bubble.
Perhaps because in this scenario, even after (only) an additional $40bn a year for the next 5 years, OpenAI will still be losing money.
I think the question is whether people are willing to pay for an LLM when there are equivalent or good-enough free competitors available.
One could argue that LLMs will change the world, but that doesn't guarantee that LLM companies will capture any of that value.
The additional rub is that the paying power users are arguably costing these companies more money than the free users.
That's what you get when you get an intern to write a research report. Or the intern getting an AI to do it
I wonder if the intern uses ChatGPT.
https://archive.ph/9b8Ae
The numbers in the article are huge. It really shows how expensive this stage of AI development has become. I’m curious how long companies can keep operating at that level of burn.
The tech giants have a lot of cash-flow to burn, and at this point it's better for their investors that they do rather than admit it was a loss or the stock value would fall.
It might be a huge pile of cards but this time they have the coffers for it, if/when it falls it will be from one of the fake it till you make it companies like openai that are built on promises (and frankly their revenue numbers are no where near their promised spend so they seem dependant on more and more deals)
When the cartelisation of the AI providers will happen, they’ll come back and demand an arm and a leg
A cartel of last entrants can create a much smaller debt problem for themselves to ruin anyone who wasted time in the industry until now.
[flagged]
That YouTube video you posted nothing to do with this article and is AI generated slop.
I'd like to think that OpenAI's digital ad revenue should reach 10% by 2030 an then accelerate from there. In my opinion, the data that ChatGPT has on a user is better than the inferred user data from Instagram/FB usage. I think ChatGPT can build a better advertisement profile of each user than Meta can which can lead to better ad targeting.
This doesn't account for OpenAI's other ambitions such as Sora app.
Hey Sam Altman or OpenAI employee, if you are reading this, I think you should buy the North American version of TikTok if the opportunity presents itself. The future of short videos will be heavily AI generated/assisted. Combine Tiktok's audience with your Sora tools and ChatGPT data and you got yourself a true Instagram competitor immediately. If the $14b sales price of US Tiktok is real, that's an absolute bargain in the grand scheme of things.
But this would fill TikTok with Sora garbage and kill it.
Tiktok has an algorithm that shows/hides content based on engagement. So any garbage wouldn't make it to the top.
From what I've seen, tons of garbage floats to the top of average non technical users. I'm not sure that Sora slop would be much worse than Daily Mail hate news slop. Or jihadist preacher slop. Or russbot misinformation slop. There's even tons of AI voiceover slop already. Sora will just be a small step towards the dead internet.
... unless someone pays Tiktok to show the garbage, because it's ads
this comment feels so eerie as I am currently reading Zuboff's "The Age of Surveillance Capitalism," which itself is interesting to read now since its written before the huge AI leap.
Also, it reminded me of the following quote, mentioned in the book, from Langdon Winner
The changes and disruptions that an evolving technology repeatedly caused in modern life were accepted as given or inevitable simply because no one bothered to ask whether there were other possibilities.
Please don't make the dystopian hypnosis box any more dystopian.
no