The one thing N. Lee Plumb knows for sure about being laid off from Amazon is that it wasn’t a failure to get on board with the company’s artificial intelligence plans.

Plumb, his team’s head of “AI enablement,” said he was so prolific in his use of Amazon’s new AI coding tool that the company flagged him as one of its top users.

Many assumed Amazon’s 16,000 corporate layoffs announced recently reflected CEO Andy Jassy’s push to “reduce our total corporate workforce as we get efficiency gains from using AI extensively across the company.”

But like other companies that have tied workforce changes to AI, it can be hard for economists to know if AI is the real reason behind the layoffs or if it’s the message a company wants to tell Wall Street.

“AI has to drive a return on investment,” said Plumb, who worked at Amazon for eight years. “When you reduce head count, you’ve demonstrated efficiency, you attract more capital, the share price goes up.”

“So you could potentially have just been bloated in the first place, reduce head count, attribute it to AI, and now you’ve got a value story,” he said.

Amazon said in an emailed statement that AI was “not the reason behind the vast majority of these reductions.”

“These changes are about continuing to strengthen our culture and teams by reducing layers, increasing ownership, and helping reduce bureaucracy to drive speed and ownership,” it said.

Whatever it was that cost Plumb his job, his skepticism about AI-driven job replacement is one shared by many economists.

“We just don’t know,” said Karan Girotra, a professor of management at Cornell University’s business school. “Not because AI isn’t great, but because it requires a lot of adjustment and most of the gains accrue to individual employees rather than to the organization. People save time and they get their work done earlier.”

If an employer works faster because of AI, Girotra said it takes time to adjust a company’s management structure in a way that would enable a smaller workforce. He’s not convinced that’s happening at Amazon, which he said is still scaling back from a glut of hiring during the COVID-19 pandemic.

A report by Goldman Sachs said AI’s overall impact on the labor market remains limited, though some effects might be felt in “specific occupations like marketing, graphic design, customer service, and especially tech.” Those are fields involving tasks that correlate with the strengths of the current crop of generative AI chatbots that can write emails and marketing pitches, produce synthetic images, answer questions, and help write code.

But the bank’s economic research division said in its most recent monthly AI adoption tracker that, since December, “very few employees were affected by corporate layoffs attributed to AI,” though the report was published Jan. 16, before Amazon, Dow, and Pinterest announced layoffs.

San Francisco-based Pinterest was the most explicit in asserting that AI drove it to cut up to 15% of its workforce. Pinterest echoed that message in a regulatory disclosure that said the company was “reallocating resources to AI-focused roles and teams that drive AI adoption and execution.”

Expedia has voiced a similar message but the 162 tech workers the travel website cut from its Seattle headquarters included several AI-specific roles, such as machine-learning scientists.

Dow’s regulatory disclosures tied its 4,500 layoffs to a new plan “utilizing AI and automation” to increase productivity and improve shareholder returns.

Cornell’s Girotra said it’s possible that increasing AI productivity is leading companies to cut middle management, but he said the reality is that those making layoff decisions “just need to cut costs and make it happen. That’s it. I don’t think they care what the reason for that is.”

Not all companies are signaling AI as a reason for cuts. Home Depot confirmed on Thursday that it was eliminating 800 roles tied to its corporate headquarters in Atlanta, though most of the affected employees worked remotely.

Home Depot’s spokesman George Lane said that Home Depot’s cuts were not driven by AI or automation but “truly about speed, agility” and serving the needs of its customers and front-line workers.

– The Associated Press contributed to this article.