Connecting the dots: Generative AI has been blamed for lots of of hundreds of layoffs over the previous yr, however proof that firms moved too rapidly to automate white-collar jobs is steadily mounting. A number of current research counsel that many employers are refilling not too long ago eradicated positions after overestimating AI’s productiveness good points and price financial savings.
In some research, roughly a 3rd of firms that tried to exchange employees with AI have both rehired a few of them or expressed remorse over the choice. The figures add to a rising physique of proof that the true price of implementing generative AI is catching companies off guard.
A late 2025 report from Forrester Analysis predicted that roughly half of AI-attributed layoffs could be quietly reversed. Nonetheless, the so-called AI boomerang impact could not profit all employees equally.
Whereas corporations may quietly rehire skilled workers, these looking for entry-level jobs should still be out of luck. Forrester additionally predicted that almost all firms will use the chance to pivot to cheaper offshore labor.
In the meantime, Gartner printed analysis in February predicting that half of the companies that eradicated customer support positions will rename and refill them by 2027. The forecast accompanied a separate October 2025 survey of 321 customer support and assist leaders, which discovered that solely 20% had really diminished headcount whereas pivoting to AI – suggesting automation has largely augmented employees relatively than changed them.
Quick Firm not too long ago lined a Robert Half report wherein about 29% of surveyed firms had rehired workers for the very same roles that they had beforehand eradicated. Finance (44%), HR (35%), and tech (32%) noticed the very best charges of rehiring, with advertising and marketing, authorized, healthcare, administrative, and buyer assist shut behind.
The corporations found that whereas AI can full many duties extra rapidly and effectively than people, high quality drops required extra human oversight than anticipated.
Of two,000 surveyed hiring managers, round 40% reported that AI couldn’t exchange institutional information, 38% stated they underestimated the necessity for human high quality management, and 35% noticed disappointing productiveness good points. These findings comply with a November report wherein knowledge from 2.4 million employees at 142 firms worldwide confirmed a current regular rise in rehirings.
Tech-sector layoffs stay at historic ranges and plenty of corporations cite AI as an element, however some are reconsidering after experiencing sticker shock. Uber burned via its complete 2026 AI coding instruments price range (spent largely on agentic platforms like Claude Code and Cursor) in simply 4 months, with its COO acknowledging it was troublesome to attach the spending to measurable enhancements.
The corporate has since capped per-employee month-to-month spending on these instruments. Rising prices have additionally prompted different AI device suppliers to tighten utilization limits.
