Visier Finds Nearly Half of U.S. Employees Exaggerate Their AI Skills at Work

Visier's survey of 1,000 U.S. full-time employees found 48% have exaggerated their AI use or expertise, while 45% feel pressured to use AI despite lacking confidence in doing so effectively. Although 54% say AI has significantly changed their role or career over the past two years, 70% are concerned about its negative career effects and only 21% strongly agree their employer provides adequate support for AI-driven workplace changes. The findings highlight a workforce-planning and training gap rather than a near-term market-moving development.

Analysis

This is a weak direct read-through for the named employers: workforce-AI adoption risk is operational rather than a near-term revenue driver. The investable implication is that reported AI-user counts and pilot announcements are increasingly poor indicators of realized productivity; investors should demand evidence in labor hours per unit, SG&A-to-sales, service levels, and retention rather than anecdotal adoption metrics. Companies that push automation without role redesign or data-governance controls risk a temporary cost increase from duplicate human/AI workflows, rework, compliance failures, and elevated attrition.

For DKS, DPZ, EBAY and F, the relevant 1-3 quarter question is whether AI spend converts into measurable labor leverage without degrading customer experience. DKS and DPZ are most exposed to frontline execution: poor training can raise fulfillment errors, store/restaurant turnover, and customer-resolution time, offsetting modest back-office savings. EBAY has greater upside from AI-assisted seller support and trust-and-safety automation, but only if fraud-loss rates and GMV conversion improve concurrently; otherwise AI features become an expense line with little multiple support.

Over 6-18 months, vendors with proprietary workflow data and embedded HR/enterprise distribution should gain share from generic copilots, as boards move from experimentation toward auditable workforce planning. However, this survey is vendor-sponsored, small, and perception-based; it does not establish that adoption anxiety translates into budget expansion. Consensus may be too quick to extrapolate broad AI productivity gains into margin estimates, particularly for regulated AMGN and EXPN, where validation, privacy, and accountability requirements slow deployment but may ultimately protect incumbents with strong governance infrastructure.

Key Decisions for Investors

• No standalone directional trade on this release; treat it as a diligence flag, not a fundamental catalyst. Require Q3/Q4 disclosures showing AI-linked labor-hour reduction, SG&A leverage, quality metrics, or retention before underwriting earnings upside.
• For a 3-6 month relative-value watch, favor long EBAY versus short a broad retail/consumer-discretionary proxy (XRT) only if seller-support automation coincides with stable fraud losses and accelerating take-rate/GMV trends. Exit if trust-and-safety expense rises faster than revenue or conversion fails to improve.
• Maintain a cautious stance on margin-expansion assumptions in DPZ and DKS through the next two earnings cycles. A material rise in labor cost per store, order-error complaints, or turnover despite AI investment would support trimming longs; evidence of labor leverage with stable NPS would invalidate the caution.
• For AMGN and EXPN, monitor 6-18 month AI productivity claims against regulated-process cycle times and compliance costs. These are potential quality beneficiaries if governance creates a deployment moat, but do not pay for the thesis until management quantifies savings and audit controls.