• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Question when will Open AI go bankrupt

when do you expect Open AI to go bankrupt

  • this year

    Votes: 3 7.3%

  • next year

    Votes: 12 29.3%

  • 2028

    Votes: 9 22.0%

  • 2029

    Votes: 4 9.8%

  • 2030

    Votes: 4 9.8%

  • 2031 to 2035

    Votes: 1 2.4%

  • 2036 to 2040

    Votes: 1 2.4%

  • 2041 to 2050

    Votes: 0 0.0%

  • beyond 2050

    Votes: 0 0.0%

  • will never go bankrupt

    Votes: 7 17.1%
  • Total voters
    41
  • This poll will close: .
23% of people are real pessimists 😀

Every company will go bankrupt eventually (or be split and sold for parts).
I voted 2029, mainly because I think Grok doesn't like OpenAI and it wishes it bad things ...
 
OpenAI will probably go bankrupt the day the AI bubble bursts. Not because they can't do great things, but because their profits can't ever realistically match their investment and inorganic growth (same as the rest of AI).

Their foundation models are reaching a limit in capability and performance despite their training and inference spending several times more power and money than their predecessor.
Despite all the promises made throughout the past 3 years, AGI isn't coming from ever-increasing LLMs and I think most people in the know are aware of this. It's only a matter of time until the markets become aware of this, too.

Either that takes a month or two years, it's going to happen regardless.
 
Last edited:
ToTTenTranz said:
OpenAI will probably go bankrupt the day the AI bubble bursts. Not because they can't do great things, but because their profits can't ever realistically match their investment and inorganic growth (same as the rest of AI).

Their foundation models are reaching a limit in capability and performance despite their training and inference spending several times more power and money than their predecessor.
Despite all the promises made throughout the past 3 years, AGI isn't coming from ever-increasing LLMs and I think most people in the know are aware of this. It's only a matter of time until the markets become aware of this, too.

Either that takes a month or two years, it's going to happen regardless.
Click to expand...
I think if the AI Bubble Bursts it will be Orcale that gets the brunt immediately which eventually knocks OpenAI out.
 
lightmanek said:
Every company will go bankrupt eventually (or be split and sold for parts).
Click to expand...
Being sold before bankruptcy isn't the same as actually going bankrupt.

Sam Altman has too many connections, OpenAI is never going away until it's not a culturally relevant brand, so at least 20 years minimum.
 
Hmm, good news if Nvidia and OpenAI are having problems. Changed my vote to '28 but it may happen sooner. What is amusing is that people are acting like the gaming industry and PC building wasn't already in a rough spot. Gaming certainly is, just going by how often sales are now. I just saw that GoG has 2077 and it expansion on sale again. Lots and lots of sales even a month after the holidays. Any holiday is being used to put products on sale.
 
marees said:
grok said 2028. so I am going with 2028
Click to expand...
Relies on Grok, which is an AI and wonders why companies are pushing it.

I kid, I kid.

All jokes aside an article about OpenAI said they LOSE money on every query, because there's a hardware connection(GPU) in there somewhere, the ones Nvidia and AMD is charging $20K dollars each for. And even if you are a premium user, you aren't really making them lots of money. And AI is plastered everywhere now, even automatic with search engines.
 
I'm betting 2028, because NVIDIA is just one of many investors. However they are burning cash like crazy and have fallen behind everyone for specific niches.

There also ALWAYS has to be a fall guy. Look at previous recessions. 2008 had a couple fall guys and the government bailed the rest out.

OpenAI made a bet on the mass consumer market wanting to pay for subscriptions to a chat bot. That was/is a bad bet. LLMs have SOME use in the professional sector. For the interpersonal sector, those that pay often have type of mental health issue and are lonely, seek validation, etc.

My current prediction is that by EOY 2028:

OpenAI will collapse in some way, or be close enough to it they will beg for US government help.

Microsoft will take a huge hit (they've already started walking back AI features in Windows and other products to minimize this, however they'll still get hit)

NVIDIA will likely see the writing on the wall, however they will still be harmed severely (If AMD could stop missing opportunities, they could profit here, alas...).

RAM will go on deep discount, and I would not rule out SK Hynix, Samsung, Micron, etc. either going bankrupt or exiting the market, depending on the company from unfulfilled contracts, negative consumer sentiment, and China's rise in the memory arena (more on that soon).

AMD will...suffer a huge loss, however they will be on better footing than NVIDIA since people need general compute and Intel can't provide it.

Intel...well...lol. They have until that time to turn stuff around, otherwise they are cooked.

Apple will be fine.

Amazon will take a huge hit.

Facebook will take a huge hit.

Google will take a a moderate to huge hit, however they will be fine.

Everyone else will try and pivot to profitability.

AI tools/LLMs won't go away. Many workplaces in tech now require them. Useless/crappy tools will go away.

This whole process will likely extend through 2030, however, Open AI's issues I think will hit hard in 2028 unless they find a way to differentiate.
 
((probably should start a related thread — "will nvidia allow open ai to go bankrupt" ))

at this point the shovel maker has full control over the gold digger
 
Mateo127350 said:
Honestly, I cant wait for the AI bubble to pop so that prices for pc building can finally be normal again.
Click to expand...
Realistic optimistic case that it pops in the Summer (because of the 18 year Economic cycle, which is 18 years since the 2008 GFC), causes prices go down by Black Friday by a lot but not mid 2025 pricing (say 32GB of DDR5 with decent speeds and latency for $200-250USD USD) and then lower Black Friday 2027 (maybe $300 for 64GB 8000 or even 8800 MT/s?) but not as low per GB as Samsung, SK Hynix and Micron go to higher speed DDR5 modules to maintain margins than producing the crap ton of lower end DDR5.
 
latest grok estimate


  1. Bankrupt by 2027 = 45% — Aligns with widespread analyst warnings of cash depletion by mid-2027 without new funding; premises amplify this by assuming investor pullback amid no ROI.
  2. Bankrupt between 2028 to 2030 = 30% — If partial funding extends runway but costs continue outpacing (e.g., $40B burn by 2028), collapse likely before projected 2029 breakeven.
  3. Bankrupt between 2031 to 2035 = 15% — Assumes prolonged survival via minimal funding or cost cuts, but tech shifts (e.g., cheaper competitors) make long-term viability low under premises.
  4. Bankrupt between 2036 to 2050 = 5% — Remote; by then, AI landscape evolves dramatically, but sustained losses without ROI would force earlier failure or acquisition.
  5. Somehow limps to sustained profitability = 5% — Least likely given premises; requires improbable breakthroughs in efficiency or monetization (e.g., ads succeeding despite user resistance) to offset $115B+ cumulative losses.

 
dangerman1337 said:
Realistic optimistic case that it pops in the Summer (because of the 18 year Economic cycle, which is 18 years since the 2008 GFC), causes prices go down by Black Friday by a lot but not mid 2025 pricing (say 32GB of DDR5 with decent speeds and latency for $200-250USD USD) and then lower Black Friday 2027 (maybe $300 for 64GB 8000 or even 8800 MT/s?) but not as low per GB as Samsung, SK Hynix and Micron go to higher speed DDR5 modules to maintain margins than producing the crap ton of lower end DDR5.
Click to expand...
Well... China would fill that void and then Samsung, SK Hynix and Micron will try to block the export of the memories overseas while are being kicked out China.
 
Well your buying Chips with money you dont have yet technically, then allocating chips that arent made yet for GPU's orders you dont have, to put in Data centers not built, to use Power from power plants not upgraded for high load, with again money that hasn't been fully vested yet because its still locked behind share holders.

Sounds like Subprime lending 2.0 anyway i look at it. Seems like people haven't learned there lesson back from 2010.
 
eek2121 said:
AI tools/LLMs won't go away. Many workplaces in tech now require them. Useless/crappy tools will go away.
Click to expand...
Economic cycles are compared to a Pendulum, because it swings from one extremes to the other. So right now is way too much AI, focus, even though it's useless, and are forced to be used. So if it ever swings back, then any use and mention of AI will be criticized to the point of extremes. Those that aren't balanced will get bankrupt in the other extreme.
aigomorla said:
Well your buying Chips with money you dont have yet technically, then allocating chips that arent made yet for GPU's orders you dont have, to put in Data centers not built, to use Power from power plants not upgraded for high load, with again money that hasn't been fully vested yet because its still locked behind share holders.

Sounds like Subprime lending 2.0 anyway i look at it. Seems like people haven't learned there lesson back from 2010.
Click to expand...
Yea it's everything all converging. Greater Depression. Like 1930's Great Depression but without the backup people had(like being able to go back to farms to at least survive).
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top