Ford laid off 5,000 workers as it deployed AI for vehicle quality inspections and design reviews. Then the company issued 152 recalls last year - the highest total in the U.S. automotive industry. It has since rehired 350 veteran engineers, including retirees, to fix persistent quality problems. Swedish fintech Klarna cut over 1,000 customer service roles after an AI assistant began handling the workload once done by 700 agents. Six months later, customer satisfaction had plummeted, and Klarna is now rehiring staff.
These reversals are not isolated. A survey of 2,000 U.S. hiring managers by recruitment firm Robert Half found that 32% had eliminated certain roles because of AI and later rehired for the same positions. Research firm Forrester reported that 55% of employers regretted AI-driven layoffs. And Gartner projects that by 2027, half of companies that reduced customer service and operational staff through AI will rehire similar roles.
Companies walk back AI cuts as reality sets in
Ford's recalled veteran engineers are now training younger employees, leading design reviews, and working to improve the AI tools themselves. "AI is an excellent tool, but it only works as well as the information it is trained on," the company said. "We misjudged that introducing AI and simply inputting requirements would result in high-quality products."
Klarna's retreat was equally sharp. CEO Sebastian Siemiatkowski said, "Our strategy focused solely on cost reduction was excessive," adding that "investing in improving the quality of human consultations is what we will do moving forward." After halting new hiring for more than a year to concentrate on AI development, the fintech is now rebuilding its human agent team so customers can speak with a person whenever they want.
HR automation hits a wall at IBM
IBM's AskHR assistant was built to replace routine human-resources tasks - leave applications, payroll management, and other repetitive work - automating 94% of those activities in AI for Human Resources functions. The company cut its HR staff by more than 200. But the chatbot could not handle the remaining 6%: ethical dilemmas, complex organizational issues, and sensitive communication problems with employees. IBM, which had considered eliminating roughly 30% of its clerical workforce, decided instead to triple its new hiring in the U.S. this year.
Gartner's analysis underscores the limits of current tools. "AI is not mature enough to fully replace the expertise, empathy, and judgment provided by humans," the firm said. The gap between automating a payroll query and resolving a harassment complaint remains wide.
Entry-level roles feel the pressure
AI is displacing some workers even as it exposes the indispensability of others. According to Stanford University's Human-Centered AI Institute "2026 AI Index" report, employment of software developers aged 22 to 25 in the U.S. fell nearly 20% compared with 2022, while hiring for developers in their 30s and 40s rose 10-15%. Employers increasingly want skilled professionals who can review AI-generated code and design entire systems, not just write routine functions.
Layoffs attributed to AI in the U.S. soared to 101,743 in the first half of this year, up from 12,742 in all of 2024, according to outplacement firm Challenger, Gray & Christmas. Entry-level and highly repetitive roles bear the brunt. Yet even Silicon Valley AI companies are paying top dollar for work only humans can do. Anthropic is hiring a "Community Lead" with a salary up to $320,000 - a role focused on fostering communication between people, not on technical development.
A tech industry insider said, "Human judgment, communication skills, and field experience - areas AI cannot replicate - are becoming even more critical," and that "as companies adopt AI, the value of talent that can verify AI outputs, handle exceptions, and persuade people will only grow."
Why this matters for HR leaders
The bounce-back in hiring after AI cuts sends a clear signal: over-automating people functions can damage customer satisfaction, product quality, and employee trust. HR leaders who view AI primarily as a staff-reduction lever risk creating holes that are expensive to fill later. Instead, the most durable strategies invest in employees who can validate AI outputs, manage exceptions, and handle the emotional and ethical dimensions of the workplace. Building a workforce with those capabilities - and figuring out where AI genuinely creates capacity rather than just cutting headcount - is now a core responsibility at the executive level. An AI Learning Path for CHROs can help leaders structure that strategic balance before the next round of tools arrives.
Your membership also unlocks:
