CEOs still want to spend on AI despite growing dissatisfaction and alarming data

Data clearly shows the world has so far seen little to no returns on investments in AI. However, a new survey suggests that the chief executives of the planet’s largest companies remain committed to the technology and plan to increase their spending on it in 2026.

According to the survey from advisory firm Teneo, the results of which are detailed in The Wall Street Journal, 68% of CEOs plan to spend even more on AI next year, despite the fact that less than half of current AI projects have generated more in returns than what they cost.

Furthermore, 84% of CEOs of large companies (those with revenue of $10 billion or more) believe meaningful returns on investments will take more than six months, and only 31% of them expect the global economy to improve any time soon.

In other words, companies will keep trying, even though doubts about the economics of it all are rising. Two-thirds of the respondents even believe that AI will increase their entry-level headcount.

Their hopes contrast with realistic analysis. Already a year ago, Daron Acemoglu, a labor economist at the Massachusetts Institute of Technology, published an estimate that productivity gains from AI will be less than 0.53% over the next ten years.

The banking and investment giant Goldman Sachs, which estimated in early 2023 that two-thirds of jobs would be exposed to automation, with 18% of global work outright replaced by it, has since stated that the technology is nowhere near where it needs to be to replace jobs at that rate.

And David Cahn at VC giant Sequoia Capital wrote already in June 2024 that AI firms need to turn something like $600 billion in revenue for the AI bet to pay off.

In a recent survey conducted by 20i, a web hosting provider, 54% of American web professionals said they were concerned about the growing reliance on AI, and 84% said they wanted more government oversight.

This might be a good idea because Wired has just reported that OpenAI is becoming more “guarded” about posting research on the economic impact of AI. The truth may be inconvenient: AI could be bad for the economy.

According to the Wired sources, the alleged censorship has frustrated at least two OpenAI employees working on its economic research team, who have quit the company.

One of them, researcher Tom Cunningham, reportedly wrote in his final farewell message that the team was forced to act like OpenAI’s propaganda arm rather than do real research.

According to the Wired sources, the alleged censorship has frustrated at least two OpenAI employees working on its economic research team, who have quit the company.

Unsurprisingly, Reddit users are having fun commenting on the results of the Teneo survey and the whole AI hype.

“It will cure cancer. Everyone will be billionaires. We will live forever, and life will be grand,” one user wrote.

“Meanwhile, on planet Earth, I can’t get AI to stop making mistakes and giving me bad answers. It’s only useful for research if the person researching knows more than the AI.”

Unlock more exclusive Cybernews content on YouTube.