Artificial intelligence is significantly changing how businesses think about their workforces in 2025. A major study by the Chartered Institute of Personnel and Development (CIPD) found that one in six employers worldwide expect AI to shrink their workforce over the next year. Specifically, 62% of these employers believe that junior management, clerical, and administrative roles are most at risk from AI automation. An even more striking finding comes from another survey showing that three in 10 companies plan to replace employees with AI, with 59% of these companies planning to replace 10% or more of their workers.

However, the picture is more complicated than just job losses. According to a massive global survey by PwC covering nearly 50,000 workers across 48 countries, workers who use AI every single day report earning higher pay, having better job security, and being more productive than those who rarely use AI. The research shows that daily AI users report 92% higher productivity compared to 58% for those who use it rarely. Yet adoption remains relatively low, with only 14% of workers using generative AI daily and just 6% using agentic AI systems daily. This gap between potential benefits and actual use suggests that companies need to focus on training workers to use AI effectively rather than replacing them.

Data from the U.S. Federal Reserve shows that AI adoption at work increased from 33.3% to 37.4% over the past year, with workers spending an average of 5.7% of their work time using AI tools. Early evidence suggests these tools are working—estimated productivity gains from AI could be around 1.3% across the entire U.S. economy. Meanwhile, companies like Meta are beginning to evaluate employee performance based on AI skills, with plans to assess workers' AI-driven impact starting in 2026. The overall message is clear: while AI is eliminating some jobs, it's also creating opportunities for those willing to learn and adapt to work alongside these powerful new tools.