Why AI-fueled layoffs will backfire

By Gautam Mukunda

RIGHT NOW there seem to be only two types of business headlines: Those dedicated to the eye-popping investments and valuations of the ever-expanding AI boom, and those chronicling a stream of layoff announcements. Strikingly, you’ll often see the same company names appearing in both.

It makes sense, I suppose. Employers in thrall to the possibilities of this powerful new technology are betting it will drive productivity — meaning fewer humans are needed. (And the post-layoff stock bump doesn’t hurt.) But ultimately, many of these cuts will likely prove unwise. In fact, they may undermine the very thing companies are so focused on: the ability to use AI to its fullest potential.

If the current downsizing is indeed a mistake, it’s one : Last month, US companies made for the last two decades. Meanwhile, many of the companies involved look healthier than ever. Amazon.com, Inc., which has announced plans to shed as many as , is enjoying share prices, while Microsoft, which is undertaking its , recently .

So if not hardship, what is driving these layoffs? In at least some cases, AI is certainly a factor. Accenture PLC, for example, announced a cut of 11,000 workers in September, declaring that these employees “.” And with AI fever sweeping corporate America, expect the technology to inspire more cuts soon. They may actually be an economic necessity: Geoffrey Hinton, the Nobel Prize-winning godfather of AI, claims that the only way they can pay off is via massive job destruction.

One problem: However promising AI tools appear, they don’t always pay off for the businesses that use them. This isn’t an argument that, as some , that the technology is useless — I’m a ChatGPT convert myself. But a Massachusetts Institute of Technology survey of 300 publicly announced corporate AI initiatives found that the executives overseeing them reported that .

When you think about it, that’s not so surprising. These tools aren’t just drop-ins that seamlessly replace workers. Most companies don’t know how to exploit their full potential — it’s not clear anyone really does. Utilizing them properly is going to require significant changes in how work is done. This is a technology that’s only a few years old and is changing by the day. With no clear roadmap to follow, companies are going to need to become more creative and innovative if they hope to adapt to an AI world and get the most out of the technology.

The current wave of job cuts is likely to make that harder. That’s because layoffs don’t just harm the people who leave — they also traumatize those who survive, hurting their . No wonder management research has also found that companies that conduct layoffs during a period of prosperity have worse than competitors who don’t reduce headcount.

What’s more, these negative effects are and rapidly-growing industries. A , for example, found that when downsizing is combined with significant changes in equipment, techniques, or processes (e.g., the type of transformation required to take advantage of AI), innovation declines because employees feel threatened and become less willing to take risks. A found that although small- and medium-sized layoffs don’t significantly hinder innovation, large downsizing does. (Such unexpected effects may be one reason .)

This isn’t to say that layoffs are all bad for the companies that do them; when organizations have too much slack, . But even there the path is fraught. If organizations are resource constrained (and given the scale of investment the AI arms race demands, even the wealthiest company could be), the effects of layoffs quickly turn negative once again.

The paradox of innovations that are as transformative as AI is supposed to be is that they are never just plug-and-play. Inventing the technology is only the first step; learning how to use it is just as hard, and just as important. It requires employees who are ready to learn, take risks, and embrace change — not ones left traumatized and fearful after their colleagues have been brushed aside. Laying off people now in anticipation of AI’s effects might seem very tempting to today’s CEOs, but most of those who do will end up regretting it.

BLOOMBERG OPINION