Trending Articles
Trending Topics
Trending
Don’t want AI to be a bubble that bursts? Then we need to talk about adoption.
That’s the message from the tech industry this week, from OpenAI blog posts to CEOs chatting onstage at Davos.
In short, if we and everyone else on the planet don’t use this technology for a wider range of tasks, it’s our own fault that the bubble bursts – that might tank the economy, but at least the price for computing kit and electricity might come down.
Over the weekend, OpenAI’s CFO Sarah Friar revealed in a blog post that her employer had again tripled its revenue, with last year’s topping $20 billion – great news, but it’s going to have to keep that pace up to pay off the many deals signed over the past year, which add up to hundreds of billions of dollars.
Friar explained that heavy upfront investment is necessary, as “at times, capacity leads usage”. So the company signs deals to get hardware when demand requires it, or as she put it: “That lets us lean forward when growth is there without locking in more of the future than the market has earned.” (Apparently $1 trillion in deals is not locking in.)
She continues: “That discipline sets up our focus for 2026: practical adoption.”
What on earth is practical adoption? An expansion of who can use AI and how they use it – so not just bored Westerners making memes, but everyone putting this technology to proper use for meaningful work.
“The priority is closing the gap between what AI now makes possible and how people, companies, and countries are using it day to day,” Friar says. “The opportunity is large and immediate, especially in health, science, and enterprise, where better intelligence translates directly into better outcomes.”
While on stage at Davos this week, Microsoft CEO Satya Nadella spoke widely about the rise of AI, but the line picked out by the Financial Times focused on the need for wider adoption to avoid the bubble bursting.
“For this not to be a bubble by definition, it requires that the benefits of this are much more evenly spread,” he said in a conversation that can be viewed on YouTube.
He later added that it would be a bubble if the only group to benefit from AI was tech companies themselves, or those in wealthy countries.
Maybe there’s a reason adoption isn’t tracking at the rate these companies would like: AI’s not delivering on what’s promised. Feel free to add a “yet” to that sentence if you’re an optimist.
A fellow Davos attendee, PwC’s global chairman, Mohamed Kande pointed to a survey by the consultancy that showed a gap between AI adoption and benefits, with just one in ten reporting cost savings and 56% said they are “nothing out of it”.
Similar research from Gartner has shown the “impact on productivity has been inconsistent,” while Workday suggested AI brings benefits, but they’re largely lost dealing with AI slop or fixing mistakes. A result echoed by a Zapier study.
And then there was that MIT report that found 95% of generative AI pilots were failing. And so on.
Kande argued that the problem is down to execution – it’s not the technology’s fault, he reportedly said, but our own dithering and fear of failure that’s holding us back.
Given the huge sums and effort spent on this round of AI, hopefully it can be put to better use than it is now – there’s got to be a better use for AI than making email somehow worse. We’d all like “practical adoption” to happen globally if it means better drugs, cures for cancer, and productivity boosts that let us work four days a week.
But the tech needs to be actually useful for AI to be fit for business let alone more serious pursuits. Maybe we’ll get better at using it. Maybe AI will improve. Either way, expect to hear the word “adoption” a lot this year.
