4 ms·

"Above $100B" is > 30x annualized revenues reportedly in excess of $3B. The company is still unprofitable, reportedly burning around $1B every month. The risk

by cs702 2y ago

"Above $100B" is > 30x annualized revenues reportedly in excess of $3B. The company is still unprofitable, reportedly burning around $1B every month. The risk for investors is vertigo-inducing high. The potential payoff is vertigo-inducing uncertain. Meanwhile, Meta is doing its darnedest to turn humongous SOTA models into commodities.

api 2y ago
I think the play for these companies was/is to erect regulatory barriers to entry by exploiting AI fear-mongering to regulate or outlaw open source or upstart competitors. That seems to be failing for the most part.
- nonethewiser 2y ago
  Meanwhile all the “safety” fears are really just reframing content moderation.
JumpCrisscross 2y ago
For a growing company, the PEG ratio [1] is more appropriate than P/E. What it asks is how many years of sustained growth brings the P/E ratio in line with population. (As is customary in finance, it's inverted from its intuitive form.) OpenAI reportedly "doubled its annualized revenue to $3.4 billion in the past six months" to June from "$1.6 billion in late 2023" [2]. That implies a CAGR of 350% [a]. $100bn is a 29x revenue multiple to that ARR. S&P 500's average is 2.9x [3]. For OpenAI to match that multiple its sales would need to grow at 350% for 3 years. That would put OpenAI near the top 100 companies in America by sales [4]. Huge. But not unprecedented. [1] https://www.investopedia.com/terms/p/pegratio.asp https://www.investopedia.com/terms/p/pegratio.asp [2] https://www.theinformation.com/articles/openais-annualized-revenue-doubles-to-3-4-billion-since-late-2023?rc=gtdonx https://www.theinformation.com/articles/openais-annualized-r... [a] (3.4 / 1.6) ^ (1 / 0.5) - 1 [3] https://www.multpl.com/s-p-500-price-to-sales https://www.multpl.com/s-p-500-price-to-sales [4] https://en.wikipedia.org/wiki/List_of_largest_companies_in_the_United_States_by_revenue https://en.wikipedia.org/wiki/List_of_largest_companies_in_t...
- lewhoo 2y ago
  Information on openai changes quickly and is very unreliable, but: https://www.maginative.com/article/openais-annualized-revenue-doubles-to-3-4-billion/ https://www.maginative.com/article/openais-annualized-revenu... This site says 3.2B of that comes from subscriptions. Omitting usage via API and assuming 1 user = 240$ annually that makes 133.3 million subscriptions. https://backlinko.com/chatgpt-stats https://backlinko.com/chatgpt-stats (says Jun. 04, 2024 but this might be outdated a bit more) Here it says openai has roughly 3.9 million subs in USA. On that basis I would assume roughly 8 million in Europe. Even if we add India and China(but they don't operate in China currently from what I gather) with numbers raised linearly along their population this doesn't add up. Maybe they rake in a lot via their enterprise plans or am I dumb and missing something ?
  - JumpCrisscross 2y ago
    > Information on openai changes quickly and is very unreliable Agree. My napkin analysis is dependent on the reliability of those data. > Omitting usage via API That's a big assumption. I'd assume larger purchasers have negotiated a minimum buy in exchange for discounts.
- cs702 2y ago
  > sales would need to grow at 350% for 3 years I assume you mean 350% per year over 3 years. That works out to sales growing (1+3.5)^3 = 91-fold over three years. Sure, it could happen. What's more, if OpenAI achieves AGI, growth could be even faster. But what are the odds? For investors counting on those crazy insane growth rates to justify a present valuation "above" $100B, the potential payoff is vertigo-inducing uncertain.
  - JumpCrisscross 2y ago
    I did my math wrong. It’s grow 350% for about 19 months. 10x.
    - cs702 2y ago
ben_w 2y ago
> The company is still unprofitable, reportedly burning around $1B every month. Where do these reports come from? I'm not saying they're wrong (or right), but I've never seen anyone link to an American equivalent of Companies House in the UK, e.g. https://find-and-update.company-information.service.gov.uk/company/12640701/filing-history https://find-and-update.company-information.service.gov.uk/c... Closest is newspapers, but I have no way to tell which are reputable in financial accounting, vs. which are rumour-mongers repeating what ultimately turns out to be someone tweeting a guess based on nothing.
ergocoder 2y ago
What Meta is doing seems borderline illegal... It basically funds a different area with ads money to destroy competition.
- JumpCrisscross 2y ago
  > funds a different area with ads money to destroy competition They’re commoditising their complement. That isn’t anticompetitive per se.
iJohnDoe 2y ago
$1B a month is almost unfathomable. I know the world does work this way, but there are so many things we could do for humanity with $1B a month.

by cs702 2y ago

api 2y ago

I think the play for these companies was/is to erect regulatory barriers to entry by exploiting AI fear-mongering to regulate or outlaw open source or upstart competitors. That seems to be failing for the most part.

nonethewiser 2y ago
Meanwhile all the “safety” fears are really just reframing content moderation.

JumpCrisscross 2y ago

For a growing company, the PEG ratio [1] is more appropriate than P/E. What it asks is how many years of sustained growth brings the P/E ratio in line with population. (As is customary in finance, it's inverted from its intuitive form.) OpenAI reportedly "doubled its annualized revenue to $3.4 billion in the past six months" to June from "$1.6 billion in late 2023" [2]. That implies a CAGR of 350% [a]. $100bn is a 29x revenue multiple to that ARR. S&P 500's average is 2.9x [3]. For OpenAI to match that multiple its sales would need to grow at 350% for 3 years. That would put OpenAI near the top 100 companies in America by sales [4]. Huge. But not unprecedented. [1] https://www.investopedia.com/terms/p/pegratio.asp https://www.investopedia.com/terms/p/pegratio.asp [2] https://www.theinformation.com/articles/openais-annualized-revenue-doubles-to-3-4-billion-since-late-2023?rc=gtdonx https://www.theinformation.com/articles/openais-annualized-r... [a] (3.4 / 1.6) ^ (1 / 0.5) - 1 [3] https://www.multpl.com/s-p-500-price-to-sales https://www.multpl.com/s-p-500-price-to-sales [4] https://en.wikipedia.org/wiki/List_of_largest_companies_in_the_United_States_by_revenue https://en.wikipedia.org/wiki/List_of_largest_companies_in_t...

lewhoo 2y ago
Information on openai changes quickly and is very unreliable, but: https://www.maginative.com/article/openais-annualized-revenue-doubles-to-3-4-billion/ https://www.maginative.com/article/openais-annualized-revenu... This site says 3.2B of that comes from subscriptions. Omitting usage via API and assuming 1 user = 240$ annually that makes 133.3 million subscriptions. https://backlinko.com/chatgpt-stats https://backlinko.com/chatgpt-stats (says Jun. 04, 2024 but this might be outdated a bit more) Here it says openai has roughly 3.9 million subs in USA. On that basis I would assume roughly 8 million in Europe. Even if we add India and China(but they don't operate in China currently from what I gather) with numbers raised linearly along their population this doesn't add up. Maybe they rake in a lot via their enterprise plans or am I dumb and missing something ?
- JumpCrisscross 2y ago
  > Information on openai changes quickly and is very unreliable Agree. My napkin analysis is dependent on the reliability of those data. > Omitting usage via API That's a big assumption. I'd assume larger purchasers have negotiated a minimum buy in exchange for discounts.
cs702 2y ago
> sales would need to grow at 350% for 3 years I assume you mean 350% per year over 3 years. That works out to sales growing (1+3.5)^3 = 91-fold over three years. Sure, it could happen. What's more, if OpenAI achieves AGI, growth could be even faster. But what are the odds? For investors counting on those crazy insane growth rates to justify a present valuation "above" $100B, the potential payoff is vertigo-inducing uncertain.
- JumpCrisscross 2y ago
  I did my math wrong. It’s grow 350% for about 19 months. 10x.
  - cs702 2y ago

ben_w 2y ago

> The company is still unprofitable, reportedly burning around $1B every month. Where do these reports come from? I'm not saying they're wrong (or right), but I've never seen anyone link to an American equivalent of Companies House in the UK, e.g. https://find-and-update.company-information.service.gov.uk/company/12640701/filing-history https://find-and-update.company-information.service.gov.uk/c... Closest is newspapers, but I have no way to tell which are reputable in financial accounting, vs. which are rumour-mongers repeating what ultimately turns out to be someone tweeting a guess based on nothing.

ergocoder 2y ago

What Meta is doing seems borderline illegal... It basically funds a different area with ads money to destroy competition.

JumpCrisscross 2y ago
> funds a different area with ads money to destroy competition They’re commoditising their complement. That isn’t anticompetitive per se.

iJohnDoe 2y ago

$1B a month is almost unfathomable. I know the world does work this way, but there are so many things we could do for humanity with $1B a month.