75% of C-suite executives are optimistic about agentic AI, while 48% still intend to implement cuts.

      A recent survey of 29 C-suite leaders from Fortune 500/1000 companies found that 75% are optimistic about agentic AI, with 52% anticipating either growth or stability in headcount. However, 48% plan to reduce their workforce, and the limited sample size does not provide a statistically significant representation.

      Three-quarters of executives at major firms believe that agentic AI either meets expectations or is underrated. This conclusion comes from a poll conducted by AI Infra Summit during an exclusive CEO event in April, attended by representatives from companies such as Amazon, Dell Technologies, FedEx, Hitachi, Lenovo, MasterCard, Mercedes-Benz, Wayfair, and Zoom.

      The optimistic overall finding reveals that 50% believe the excitement around agentic AI is warranted, 25% think it is underestimated, and only 25% feel it is overestimated. However, the implications for employment are less clear, and given the small sample size of 29, individual responses can significantly affect reported percentages.

      Regarding headcount projections over the next three years, executives showed a nearly even split. While 52% anticipate team expansion (14% believing in direct growth and 38% in maintaining output at existing levels), 48% foresee reductions, with 10% expecting to replace many roles with AI agents and 38% planning to decrease headcount via AI automation.

      These results align with broader trends. A Mercer survey of 825 C-suite leaders indicated that almost all expect some AI-related workforce reduction within the next two years. Additionally, a Chief Executive survey revealed that one-third of companies are already utilizing AI to downsize, despite 53% planning to expand their workforce. Fortune noted that 66% of CEOs are pausing hiring while boosting AI investments. The overall signals are mixed: corporate enthusiasm for AI contrasts with a cautious stance on staffing.

      Ed Nelson, co-founder of AI Infra Summit, commented that discussions among leaders have evolved beyond whether AI functions. “The shared view was that high-level conversations have shifted from questioning AI's efficacy to exploring effective utilization of agents,” he explained. “Some leaders noted that AI has already saved their organizations hundreds of millions of dollars.”

      The survey revealed that 95% of participants believe AI has progressed past the pilot phase in their organizations, with 40% stating that it is embedded in their core products and strategies. This aligns with broader industry data, as Google reported at Cloud Next 2026 that 89% of business teams are currently using AI agents, with an average of 12 AI implementations per organization.

      Regarding budgets, 36% indicated that traditional IT spending is being redirected to finance AI investments, while another 46% are establishing additional budgets specifically for AI. A blended model approach was also prevalent: 85% are utilizing a combination of third-party foundation models from providers like Microsoft, OpenAI, and Google, alongside proprietary layers they have developed.

      Executives also discussed which roles are likely to increase as AI takes over more tasks. The general consensus indicated that generalists—those with lateral thinking, analytical skills, and the ability to connect ideas across different areas—will be favored, whereas deep specialization may dwindle as AI absorbs specialized knowledge.

      “Integrating AI represents a human resources challenge, but it should not simply be viewed as an upskilling initiative,” Nelson asserted. “It poses a significant operational challenge. It’s not merely about applying AI to flawed processes.”

      Cautions regarding the findings include the limited sample size of 29 respondents, which is too small to accurately represent Fortune 500 or 1000 companies. These results should be regarded as a directional insight into the feelings of a self-selected group of senior AI decision-makers rather than as a definitive measure of company-wide behavior.

      Additionally, the event was hosted by a company that sells tickets to AI infrastructure conferences, which may bias the framing towards optimism. Larger surveys from PwC, Grant Thornton, and Mercer, with sample sizes in the hundreds or thousands, present a more nuanced picture. A 2026 enterprise AI report revealed that 79% of organizations encounter challenges with AI adoption, a notable increase from 2025, and 54% of C-suite executives acknowledged that AI adoption is causing significant disruption within their companies.

      The disparity between executive enthusiasm and organizational reality remains a key characteristic of enterprise AI in 2026. While leaders affirm that AI is effective, they also contend that their organizations are not equipped to utilize it effectively. These two statements are not contradictory but highlight why uncertainties about job impacts persist even as the technology advances.