Organizations forced to rehire managers following AI-driven layoffs

• Key Insight: Discover how replacing managers with AI prompts costly rehiring and capability gaps.  
• What's at Stake: Declining manager trust could erode productivity, retention, and competitive resilience.  
• Supporting Data: 32.7% of firms rehired 25–50% of roles they initially eliminated for AI.  
  Source: Bullets generated by AI with editorial review

AI has proven capable of handling many functions within an organization, but it still can't replace effective management — and companies risk significant challenges by trying.

Among organizations that conducted AI-driven layoffs, a new study from HR solutions platform Careerminds revealed that 32.7% have already rehired between 25% and 50% of the roles they initially eliminated. For those that eliminated roles altogether to make room for AI, 35.6% said they had to bring back more than half of those positions. Most affected by these decisions are managers, whose jobs are often the first to go despite their importance. 

"People have taken out large portions of management and replaced it with AI functions thinking that it could make those leaders more productive and manage larger teams at a fraction of the cost," said Doug Dennerline, CEO of performance management company Betterworks. "But it's a mistake that's probably going to cause much bigger problems."

Read more: AI is reshaping jobs, not replacing them

Already several major tech companies have reduced management layers as they invest more heavily in AI. Amazon eliminated roughly 14,000 corporate jobs in 2025 — nearly 4% of its white-collar workforce — primarily targeting early-career employees and mid-level managers to help fund AI initiatives. Meta recently cut thousands of positions, with software engineering managers and other mid-level roles accounting for about one-third of the reductions. Meanwhile, Coinbase laid off 14% of its workforce and moved away from "pure manager" roles in favor of teams where individual contributors work more directly alongside AI agents.

The problem with this strategy, and the reason it's leading to so many rehires, according to Dennerline, is that traditionally, managers support up to 10 employees each. Significantly altering that ratio could weaken manager-employee relationships, leading to lower engagement and reduced organizational effectiveness.

"It's the most critical and hardest job to do in any organization," Dennerline said. "But when done well, employees that work under managers that know how to give feedback in order to grow them, not to criticize them have the best engagement and the best results — and [AI can't replicate that]."

Coexisting with AI

First, Dennerline urged organizations to wait a period after integrating new AI tools to gauge their abilities before sacrificing headcount for greater AI spending. In many cases, there may be a way for existing managers to be relocated or retrained to work alongside those tools. If companies really need to reduce management layers to cut costs and afford those AI integrations, Dennerline said, they need to equip remaining managers with AI tools that help them effectively support larger teams. This could include programs that help them sort and organize responsibilities, respond to emails, formulate helpful and bias-free feedback, or manage their schedules.  

Read more: 5 things to consider so your AI adoption doesn't fail you

"They have to be smart about this," Dennerline said. "They have to give up on the notion that they can just whack out a couple layers and everything's going to be okay, not unless they have enabled AI to be the mechanism that helps them reach more people effectively."

Despite many organizations attempting to right their wrongs by rehiring for those roles and reinstating certain positions, in most cases the damage is already done and the trust managers had in leadership is broken. Losing that relationship could jeopardize the longevity of any organization if they can't convince qualified talent to return. 

"At the end of the day [AI] is still just a piece of technology," Dennerline said. "It's important that what technology is doing and how it's working represents what an organization really needs, but that requires human oversight and always will."