Opinion I am so sick and tired of AI hype. I'm not the only one.
Yeah, with the release of ChatGPT-3.5 in November 2022, AI became a huge deal. Now, though, the AI revolution, once heralded as the next big leap for businesses worldwide, is facing a sobering reality check. Recent data reveals a marked slowdown in AI adoption.
The simple truth is that AI is not living up to its rep. Sure, Elon Musk still thinks "there will come a point where no job is needed." Especially in the US government, where his Department of Government Efficiency (DOGE) is using AI to help fire employees and, eventually, replace them.
Other people, however, have figured out that AI is not ready for prime job time. AI adoption rates in business are stalling. In the US, general AI adoption rates are grinding to a halt. According to the Fall 2024 Slack Workforce Index, AI adoption growth rates among US workers have slowed over the last three months to a mere percentage point gain versus near-double digit gains in the same period a year earlier.
Why? Well – surprise, surprise! – they don't know how to use it. Christina Janzer, Slack's senior VP of research and analytics, explained: "Too much of the burden has been put on workers to figure out how to use AI. To ensure adoption of the technology, it’s important that leaders not only train workers but encourage employees to talk about it and experiment with AI out in the open.”
People are experimenting with AI, though often with AI sexting rather than anything that affects a business's bottom line. I find it interesting, though, that neither workers nor executives know how to use AI for the job.
A recent IDC research paper found that while IT decision-makers expect to triple their AI investments in 2025, there's still a disconnect between their spending and their doubts about AI's value. The study found that "37 percent of management remain skeptical or have reservations toward AI."
Indeed, 45 percent of companies rank AI implementation as a significant challenge, with 38 percent citing integration issues as a primary concern. In other words, they still don't know what they're doing, and it shows. As Microsoft CEO Satya Nadella recently observed, there's still no killer app for AI and we still don't understand how to use AI effectively in business. This, mind you, is from someone who's invested over 10 billion bucks in AI.
If you look closely at what Microsoft has been doing with AI, you'll see that it's pulling back in places some of us wouldn't think to look. Microsoft has canceled more than a gigawatt of datacenter operations in addition to numerous 100-plus MW agreements.
As Edward Zitron, CEO of tech PR biz EZPR, points out: "These numbers heavily suggest that Microsoft – the biggest purchaser of Nvidia's GPUs and, according to investment bank TD Cowen, 'the most active [datacenter] lessee of capacity in 2023 and 1H24' – does not believe there is future growth in generative AI, nor does it have faith in (nor does it want responsibility for) the future of OpenAI."
Less electrical juice means fewer datacenters and fewer AI resources, which means Microsoft isn't seeing as much AI demand as it thought it would.
Why not? Well, Canalys reckons AI licenses aren't selling as much as they'd hoped. Is it too expensive, or does it lack value or a killer app? I'd say both.
At the recent Canalys Channel Forum in Berlin, Alistair Edwards, Canalys chief analyst, said: "Almost every customer is struggling to really define how they can deploy AI effectively internally, how they can drive change within their business organizations and processes, and how they can build the models cost effectively to deliver the returns they need."
Simultaneously, AI continues to deliver plausible, but wrong, answers to questions. The BBC, for example, found that when asked to summarize news stories, over half the time major chatbots make blunders in their responses. For example, both ChatGPT and Copilot insisted Rishi Sunak and Nicola Sturgeon were still in office after they had exited.
Worse still, according to a Nature study, the bigger, "better" LLMs tend to be the ones that deliver the worst answers. I've been noticing this myself. I use Perplexity now as my main search engine, but in the last few weeks alone I've found its answers to be less trustworthy. This is ticking me off.
What about Microsoft Copilot? Isn't it going great guns? It appears not. An October Gartner survey found few companies have moved Copilot initiatives beyond the pilot stage. In particular, they're finding tangible business impact to be elusive. In addition, implementation and security measures are requiring more effort than anticipated.
More recently, in a Gartner discussion, a government IT executive noted: "Copilot is so far behind that it is frustrating to use, makes frequent mistakes, and doesn't really have the omniscient integration across the Azure, Office 365, Sharepoint Teams space that it should to be effective at good use cases."
He concluded: "On the other hand, it's really great for preparing meeting minutes and for skipping meetings because of the lovely summaries. But that makes it a one-trick pony. And by the way, why have meetings if they get skipped?"
Good question.
A major trend is that AI's primary business use case is to eliminate make-work, such as most meetings. One especially amusing use of this is OPM Reply. This AI-driven tool enables employees to use AI to answer DOGE's ultimate micromanagement demand of asking everyone for five-point emails on what they did last week. We have a perfect ouroboros of AI-driven pointless work since DOGE is believed to use AI to read these messages.
Pointless is increasingly becoming the word I use for AI. Yes, it can be helpful when used carefully as a tool, but that's not what I see happening. Instead, AI's being used as either a lazy way to create second-rate work or to make work. Businesses are also finally figuring this out. In Gartner Hype Cycle terms, we're entering the Trough of Disillusionment. I'll wake you up when we start climbing the Slope of Enlightenment. ®