Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/when-would-ai-application-business-become-profitable.21797/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021770
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

When would AI application business become profitable?

tonyget

Well-known member
The AI industry as hot as it is now, only the hardware business is making money,ie Nvidia TSMC broadcom etc. The actual AI application business IS NOT making money,OpenAI still loss tens of billions of dollars every year,and people predict that is won't change before 2030. Other players such as Google Microsoft etc,still need revenues generated from other department to subsdise their AI business.

At the moment the AI application business is burning investors money,of course investors are still enthusiastic about AI industry as of now. But I don't think this enthusiasm will lasts forever,after all,a health industry needs to generate profit in order to be sustainable
 
The AI industry as hot as it is now, only the hardware business is making money, ie Nvidia, TSMC, Broadcom, etc. The actual AI application business IS NOT making money, OpenAI still loss tens of billions of dollars every year, and people predict that is won't change before 2030. Other players such as Google Microsoft etc, still need revenues generated from other department to subsidize their AI business.

At the moment the AI application business is burning investors money, of course investors are still enthusiastic about AI industry as of now. But I don't think this enthusiasm will lasts forever, after all, a health industry needs to generate profit in order to be sustainable

OpenAI was originally structured as a non profit so I don't think that is a good example. AI is already making/saving companies money, we are just not seeing it. AI is disrupting search in a big way. AI is cutting headcount in most industries. AI is going to hit the chip design business hard. I agree, give it another 5 years and it will be hard to find industries that AI has not disrupted.

Just like the gold rush and Dotcom bust, the companies selling infrastructure are getting rich. TSMC credits AI to much of its growth this year, which is true, but non AI chips are getting sucked into AI infrastructure. Broadcom is an AI chip company now? My iPhone is AI now? AI PCs? This is déjà vu the DotCom era. Non Dotcom companies became Dotcom companies then came the bust.
 
I guess the concern now is not so much if AI can be profitable but more on the rate of returns. If the rate of return is only say 5% on capex then perhaps the current rate of capex will face significant pressure to continue at the same pace.
 
I guess the concern now is not so much if AI can be profitable but more on the rate of returns. If the rate of return is only say 5% on capex then perhaps the current rate of capex will face significant pressure to continue at the same pace.

AI seems a bit like crypto mining. What was the rate of return on crypto server farms?
 
AI seems a bit like crypto mining. What was the rate of return on crypto server farms?
Honestly I don't know. I'm never into crypto. However, I understand in those days, compliance/IT is going around to check if employees are secretly mining crypto using office electrical sockets after noticing electrical bills jump significantly. I guess that might means crypto mining might not worth the effort if one needs to pay their own electrical bills. 😂
 
There are several ways to consider the benefits that AI can bring to the real world. The first is, undoubtedly, profit. However, for many individuals and corporations, reducing costs and improving efficiency, productivity, and competitiveness are significant advantages that AI can provide. While these factors may not generate immediate profits, they often play a critical role in determining a company’s strength and long-term survival.
 
OpenAI was originally structured as a non profit so I don't think that is a good example. AI is already making/saving companies money, we are just not seeing it. AI is disrupting search in a big way. AI is cutting headcount in most industries. AI is going to hit the chip design business hard. I agree, give it another 5 years and it will be hard to find industries that AI has not disrupted.

Just like the gold rush and Dotcom bust, the companies selling infrastructure are getting rich. TSMC credits AI to much of its growth this year, which is true, but non AI chips are getting sucked into AI infrastructure. Broadcom is an AI chip company now? My iPhone is AI now? AI PCs? This is déjà vu the DotCom era. Non Dotcom companies became Dotcom companies then came the bust.
For sure, the infrastructure suppliers are getting rich ... and the AWS/Microsoft/Googles are seeing efficiency breakthroughs which will be seen through departments being downsized AND getting more productive at the same time ... the Mag 7 companies all have deep pockets to expand their cloud services / compute power which is a huge driver of the business cycle atm

... however, for the average joe the AI PC and AI phone are being pumped up in a media frenzy atm. Not all people will be tech savvy enough to be able to leverage AI / LLM's to become more efficient and productive, so they wont see alot of extra value in upgrading their PC. I think there will be a pause when the Mag 7 decide they have enough cloud resources / compute power and cut back on their CapEx, which will happen ... and we return to the typical Semi Conductor business cycle
 
AI profitability is clearly a game of patience. It’s burning investor cash right now, but tons of value is being created in the form of cost savings, efficiency and productivity. Meanwhile, hardware companies like TSMC and even Nvidia are cashing in, while the AI's biz in the real world is still figuring out how to turn a profit. Even the excitement may last only for a time, but the industry will have to prove sustainable returns to avoid following the dot.com bust.
 
AI is already making/saving companies money, we are just not seeing it
Here is some surveys

Chart.jpg

Source : page 19
 
Thanks. Anyway to know the actual ROI these companies are seeing ?
That might be a moot question, similar to asking “what was the ROI of a company adopting internet use and commerce ?” The real question might be, who wins and looses big from the Gen AI disruption. Internet adoption birthed Netflix and killed off Blockbuster, even though Netflix is far different from the DVD rental service they started as. Google and Facebook, plus others helped create whole new categories of media, while vastly reshaping all of advertising, where a lot of traditional advertising was displaced. And then there were failed startups like Webvan that had the basic idea right (ish), but couldn’t draft on the bigger economy like the much later and successful arrival, Instacart.

So who wins and loses big from the disruption ? And are the new unicorns of the day, like OpenAI, Anthropic, HuggingFace and Perplexity closest to a Yahoo, or a Google, a MySpace or a Facebook ? Or are we missing the real future startup winners ?
 
Here is some surveys
This is great info!

Just a comment - that this data is hard to really interpret unfortunately. A company that I used to work at:

- Implemented AI, but moved the goal posts several times for what would constitute success as they found limitations with the product (in this case, security limitations and changes). This is a facet of being an early mover on evolving tech.

- The company is compelled to show success with AI internally, as it can help to win contract bids externally. i.e. they are incentivized to be biased towards 'it exceeded our expectations, and we've mastered it'

- Like all companies, how many executives will admit they invested millions in something and it failed. Their bonus is on the line for 'good performance', so the humans running these places are incentivized to declare success too.
 
That might be a moot question, similar to asking “what was the ROI of a company adopting internet use and commerce ?” The real question might be, who wins and looses big from the Gen AI disruption. Internet adoption birthed Netflix and killed off Blockbuster, even though Netflix is far different from the DVD rental service they started as. Google and Facebook, plus others helped create whole new categories of media, while vastly reshaping all of advertising, where a lot of traditional advertising was displaced. And then there were failed startups like Webvan that had the basic idea right (ish), but couldn’t draft on the bigger economy like the much later and successful arrival, Instacart.

So who wins and loses big from the disruption ? And are the new unicorns of the day, like OpenAI, Anthropic, HuggingFace and Perplexity closest to a Yahoo, or a Google, a MySpace or a Facebook ? Or are we missing the real future startup winners ?
Good point. But the respondents of the survey said they are satisfied with the ROI. That’s very vague so I wonder what exactly is the ROI.
 
Back
Top