Forgot your password?
Close
typodupeerror
Businesses AI

Majority of CEOs Report Zero Payoff From AI Splurge 53

Posted by BeauHD from the pending-ROI dept.
A PwC survey of more than 4,500 CEOs found that over half report no revenue growth or cost savings from their AI investments so far, despite massive spending. Of the 4,454 business leaders surveyed, only 12% saw both lower costs and higher revenue, while 56% saw neither benefit. "26% saw reduced costs, but nearly as many experienced cost increases," adds The Register. From the report: AI adoption remains limited. Even in top use cases like demand generation (22 percent), support services (20 percent), and product development (19 percent), only a minority are deploying AI extensively. Last year, a separate PwC study found that only 14 percent of workers indicated they were using generative AI daily in their work. Despite the CEOs' repsonses, PwC concludes more investment is required. It claims that "isolated, tactical AI projects" often don't deliver measurable value, and that tangible returns instead come from enterprise-wide deployments consistent with business strategy. [...]

In terms of the broader picture, PwC says it found CEO confidence has hit a five-year low, with only 30 percent optimistic about revenue growth (down from 38 percent last year). This points to growing geopolitical risk and intensifying cyber threats, as well as uncertainty over the benefits and downsides of AI. Unsurprisingly, concern remains over tariffs as the Trump administration continues its erratic approach to policy, with almost a third of company chiefs saying tariffs are expected to reduce their company's profit margin in the year ahead. In the U.S., 22 percent indicate their corporation is highly or extremely exposed to tariffs. PwC warns that companies avoiding major investments due to geopolitical uncertainty underperform peers by two percentage points in growth and three points in profit margins.
This discussion has been archived. No new comments can be posted.

Majority of CEOs Report Zero Payoff From AI Splurge More

Majority of CEOs Report Zero Payoff From AI Splurge

Comments Filter:

  • Other uses for AI? (Score:5, Insightful)

    by SoCalChris ( 573049 ) on Tuesday January 20, 2026 @06:42PM (#65938350) Journal

    Maybe slashdot editors could use AI to scan for duplicate stories that are on the homepage at the same time?

  • Keep trying. We've got to get these figures looking good, or our investments are bust !
  • So they resent and despise you. The ruling class really really hate consumers and the working class.

    They could care less if AI is profitable because the goal isn't to be profitable it's to get rid of workers and consumers but still have private jets and yachts.

    It's a return to feudalism only you don't even get the affordances of a peasant.
  • But they didn't quite manage the ultimate editor fail of having the dupe consecutive to the first story. Keep trying editors, we know you can do it!

  • So close. So very close. (Score:5, Funny)

    by nightflameauto ( 6607976 ) on Tuesday January 20, 2026 @06:55PM (#65938374)

    One story between the repost. Come on, editors. You can do it. You can dupe directly in a row.

  • News for Nerds (Score:1)

    by Anonymous Coward

    Twice if it Matters

  • AI is just a plot by Big Hardware to sell you more hardware.

    • Re: (Score:2)

      by ebunga ( 95613 )

      It's a plot by big hardware to sell hardware now for future use for future money. And they'll finally be able to use it just as soon as they get the go-ahead to build the datacenters which are contingent on the availability of power. The permits for the new power plants are three to five years away, and construction, assuming everything goes to plan, will begin in 2030 at the earliest, assuming the AI companies talked to the power companies five years ago.

  • Of the 4,454 business leaders surveyed, only 12% saw both lower costs and higher revenue,

    Tell me more about this 12%

    • Tell me more about this 12%

      Corporate America doesn't look kindly on "well, we tried, but it didn't go as planned, let's do something else instead" Every exec that pitches an initiative has to make it succeed. If you're a VP and you suggest, AI, you need to justify not getting fired from your job that you're overpaid to do. So yeah, I am sure this 12% is self-reported. Even if LLMs were as good as promised, it's not easy to quantify productivity gains.

      I am confident giving every employee huge quantities of super-meth would help

      • You'd think that this would move NVDA's stock price but people are too busy taking AI meth hype to notice the difference.

    • Re: (Score:2)

      by zlives ( 2009072 )

      and ask them how cloud cost savings are going for them

  • Early adopters of immature tech always see this

  • If LLMs were legit, the world would be like Roblox (Score:5, Insightful)

    by Somervillain ( 4719341 ) on Tuesday January 20, 2026 @07:09PM (#65938406)
    The promise of Generative AI is that with clever prompting, ChatGPT and others can replicate what used to take skilled professionals much much longer. If it actually worked as promised, the world would be filled with countless startups who vibe-coded killer apps to glory, disrupting established players...the way Netflix disrupted Blockbuster or Uber did for cabs. We have a great example of this: Roblox. For those without young kids, Roblox is a platform to play mostly shitty games with your friends. However, what is unique and special about it is Roblox Studio. It makes it VERY trivial to create your own game. You can clone someone else's game and customize it. Why do I bring this up?

    When you log in, you see a few huge hit games and 1000s of low-effort clones. You see a lot of sloppy mashups created by kids or some just fucking pointless games. My kids LOVE them. One was where you could log in and dress up like your favorite Disney Channel characters...and that's it...no point...my daughter loved it, I was bored out of my skull. Why do I bring this up? Some weird kid said...hey, I want to pretend I am in a generic city and look like Addison and Zed while my favorite songs of theirs plays in the background....and she/he made it so. You see brawlers where the hulk beats up superman, omniman and homelander. There are COUNTLESS low-effort "obbys" that mostly suck. Additionally, there are quite a few great games on there.

    Why are we not seeing the same with LLMs and AI. Why are we flooded with AI companies telling us about their models, but not a wave of low-effort startups? Why aren't tiny VC-funded companies upending existing markets? Afterall, Oracle makes an obscene amount with their offerings...why isn't some 5 person startup offering a copycat vibe-coded offering for a fraction of the price? Why aren't vibe coders resurrecting obsolete software and games? Why aren't we seeing a billion low-effort AI clones of popular old games? There are infinite ways to make unfathomable money with AI....but for now, all we get are AI services and promises?

    Roblox Studio does everything for shitty game development that Sam Altman, Jensen Huang and Kara Swisher say AI can do for pretty much anything and everything. Given that it's been out for 5 years, why are we not flooded with these startups producing mostly crap? Nearly EVERYONE has used Gemini or ChatGPT at least once. Very few come back regularly.

    It's no surprise that these AI investments are not paying off. If hyper-motivated entrepreneurs, usually in their early 20s, cannot take this magic tool and produce a business worth attracting VCs, how will a major corporation that actually cares about existing customers and it's reputation?

    • Re: (Score:2)

      by ebunga ( 95613 )

      Oracle would never let an LLM be trained on their valuable intellectual property.

      • To be honest, there's nothing spectacular about Oracle's products. Plenty of SQL databases are on the same level as Oracle DB, including at least one FOSS one (PostgreSQL.) Plenty of languages compete with Java, and many, even in the same field (C#) are better.

        Oracle's money is made from sales to companies scared of picking anything but the big names, and lawsuits over misleadingly written or poorly understood contracts and licenses.

        I think the GP's problem there is that he underestimates, not overestimates

    • Re: (Score:2, Insightful)

      by TheStatsMan ( 1763322 )

      The whole premise of this comment is wrong. Where are the "vibe coded killer apps?" Why did Netflix oust blockbuster? It had nothing to do with coding. Netflix ousted blockbuster because they didn't have to own real estate.

      >why are we not flooded with these startups producing mostly crap

      Have you even been to the app store lately? It's full to the brim of useless trash and clones.

      Have you been on any social media website? It's 100% flooded with useless crap. I think you've kinda missed the entire p

      • Why did Netflix oust blockbuster? It had nothing to do with coding. Netflix ousted blockbuster because they didn't have to own real estate.

        Anyone who invested in real estate back when Blockbuster was around is probably a billionaire today.

      • Re: (Score:2)

        by allo ( 1728082 )

        The post is his selection bias.

      • Have you even been to the app store lately? It's full to the brim of useless trash and clones.

        Nothing new there then.

      • "Have you even been to the app store lately? It's full to the brim of useless trash and clones."

        AI generated crap is cheaper to produce than bespoke artisanal crap. The point is you should be able to realize a better gain from the AI crap even though it is still crap, but there are few if any examples..

    • I think the big issue is that replacing a white collar worker's core tasks really does require full-blown AGI. There is no "it can almost report sales tax to the state comptroller." It either does it, flawlessly, every single time, or bust. Current reasoning models are pretty good, but they're not able to maintain focus and iteratively refine a piece of work over the course of days, which is what's needed to complete certain tasks. However, for some jobs that require content creation, it can be a big time s

      • Re: (Score:2)

        by MikeS2k ( 589190 )

        Yes if it was worth the trillions of dollars put into it, why aren't we seeing trillions worth of results. Why can't we put these "agents" to work programming an alternative to the MS/Google ecosystem for EU-Linux for example? That could be worth hundreds of millions alone.

        In work I'll do timetabling for a school district - Imagine I say to an AI Agent - on \\server\share are the building plans, student list, list of subjects, and which students chose those subjects. Make a timetable for next school y

    • Re: (Score:3)

      by Kokuyo ( 549451 )

      It is unconceivable to me how someone can actually go have a look at Roblox and still let their kids on that platform.

      "They love it!"? Well, yeah my kids used to sit in front of Netflix like little consumer zombies because they were unable to resist Netflix' recommendations and couldn't tell when an episode ended...

      So... I unsubscribed from Netflix.

      Children would love Cocaine, too, is they were given access to it.

    • Yeah here comes the boom, just like the dotcom crash, almost identical, BUT much bigger, more fallout, so bad.

      • The companies that will be gone are the ones that mostly don't matter. Most of big tech, with the exception of AWS, is made up of companies that literally contribute nothing to the world.

        • Re: (Score:2)

          by allo ( 1728082 )

          Big tech is not affect by such bubbles. Look how many big tech companies from before the dotcom bubble are still around. All of them.
          Does anybody believe OpenAI would be affected by some bubble popping? Google? Anthropic? Meta?

          • The big companies survive, not all companies do. Here is a history lesson from 00's https://en.wikipedia.org/wiki/... [wikipedia.org]
            And that isn't true that some of these bigger companies couldn't be damaged by this, they are spending multiples on AI of their yearly revenue. At some point they are going to have to pay the loans and investors off. If their revenue doesn't go up in multiples then solvency will be an issue.

            • Re: (Score:2)

              by allo ( 1728082 )

              Just thing about Facebook. People laughed how it should ever make profit. Think about Amazon or Netflix. Investors of such companies are thinking in decades. They know OpenAI is losing money now, but they know they will probably have their letter in GAFAMO and be one of their best investment. Even if they take a hit by a bubble they'll come up again. Their main asset is not even ChatGPT, but hundreds of companies using their paid API. Which isn't even cheap compares to alternatives.

  • "pilot projects are by their very nature typically small scale and isolated in order to demonstrate the viability of a concept before risking an enterprise-wide rollout."

    Trying some things out on a small scale is exactly what you want to do initially with this tech. And that being the case, it isn't realistic to expect reduced costs and higher revenue at this very early stage. Also they will find that people need training in order to understand how to work with AI effectively. That takes time.

  • Is that AI might make the jobs of their employees easier, even if it doesn't help the bottom line.

  • It creates more problems than fixes them.

Slashdot Top Deals

Your good nature will bring you unbounded happiness.

Close