Companies that rushed to replace workers with artificial intelligence (AI) are now reversing course. New research from the Federal Reserve Bank of Atlanta (Atlanta Fed) shows that most companies have yet to record clear productivity gains from this technology. These findings come at a time when employers in the United States continue to cite AI as a reason for layoffs, and several major companies are changing their plans to replace workers with machines. Meta, Microsoft, McKinsey, and Walmart are among the major organizations currently reviewing how they deploy AI in the workplace. These moves highlight a question that adoption data has not yet resolved: whether AI is truly delivering the productivity gains often cited to justify headcount reductions.

This pullback was pointed out in a November 7 post by the market-focused account unusual_whales:

"The AI layoff gamble is backfiring. Companies that rushed to replace workers with AI are now cleaning up the mess, per The Guardian. Meta scaled back plans to cut staff and use more AI agents after employees spent growing time cleaning up the agents' technical and security…"

— unusual_whales (@unusual_whales) November 7, 2026

An investigation by The Guardian in October revealed that companies that adopted AI early are facing hidden costs. Employees are spending more and more time fixing mistakes made by AI systems, some of which involve security issues.

Meta, Microsoft, and McKinsey Review Their AI Plans

Meta has scaled back some of its plans to replace staff with AI agents after discovering that workers are spending more time fixing technical and security errors generated by these systems. The company is also rebuilding part of its management structure following previous cuts to middle management. According to The Guardian, Meta has stopped using AI adoption as a factor in employee performance reviews, though the company has not disclosed how long these changes will last.

Microsoft is also facing its own challenges with AI-generated output. Executives have warned about the increasing amount of low-quality, AI-generated material circulating within the company. Ryan Roslansky, a senior Microsoft executive, criticized situations where one employee uses AI to write a document, only for a colleague to use AI to summarize it.

McKinsey and Microsoft have reportedly placed restrictions on some employee AI tools, reportedly to manage rising costs. Meanwhile, Walmart has faced complaints from staff regarding its AI-powered scheduling and task management systems, with workers pointing out unrealistic targets, inaccurate alerts, and errors stemming from automated decisions.

Fed Research Shows Limited Gains

Despite the restructuring push, federal research suggests that the promised efficiency gains have not yet been realized across the board. A March 2026 working paper by the Atlanta Fed examined approximately 6,000 companies in the US, UK, Germany, and Australia. Of these, about 70% were already using AI. Nearly 90% of surveyed executives said that AI has not brought measurable changes to productivity or employment levels over the past three years.

However, the surveyed companies expect AI to boost productivity by about 1.4% over the next three years, a projection that provides a concrete benchmark for the current corporate pullback.

A separate paper from the Atlanta Fed did identify productivity gains in certain sectors, with finance and professional services showing the clearest improvements. However, across the broader economy, these gains have been uneven.

Over 121,000 Job Cuts in the US Mention AI

This uncertainty has not slowed down headcount reductions. According to a report released by Challenger, Gray & Christmas on November 6, there were 121,401 AI-related job cut announcements in the first ten months of 2026, accounting for about 20% of all announced layoffs this year.

Challenger cautioned that a company citing AI does not prove that the technology actually replaced those workers. Some companies may be restructuring, cutting costs, or shifting funds to AI infrastructure.

This gap—massive AI-related layoffs while executive surveys show little actual change—makes the recent corporate reversals more than just isolated individual company decisions.

The firm also reports that deteriorating market and economic conditions have surpassed AI as the primary reason for layoffs. According to Challenger, job cuts in October fell to their lowest level in three years.

Research from the Atlanta Fed notes that AI is still being deployed in fields such as software development, data analysis, and customer support. The latest findings are based on executive surveys across four countries. From here, the reference points are clear: Challenger's monthly AI-related job cut statistics, the approximately 1.4% productivity gain projected by companies for the Atlanta Fed over the next three years, and whether more employers will follow Meta in decoupling AI usage from performance reviews.