Half of Companies Could Already Be Using AI in Your Performance Reviews

Summary

This article addresses the accelerating integration of artificial intelligence into employee performance review processes across U.S. and U.K. organizations. Drawing on a General Assembly survey of over 500 senior leaders, the article argues that AI adoption in HR functions — including performance evaluations — is already widespread, with 47 percent of respondents reporting AI usage as a factor in performance reviews. Key evidence includes statistics on leadership encouragement of AI (93 percent), regular AI usage among leaders (82 percent), and a separate finding that only 30 percent of HR professionals have received job-specific AI training. The article presents perspectives from General Assembly CEO Daniele Grassi and HR consultant Bryan Driscoll, capturing both employer rationale — efficiency, standardization, and data-driven reporting — and worker and expert concerns around algorithmic bias, lack of transparency, and over-reliance on automation. The implied conclusion is that AI's role in performance management will expand, with outcomes heavily dependent on the quality of manager training and the establishment of ethical usage guidelines.

Strengths and limitations

Strengths: The article draws on a quantified survey dataset (500+ senior leaders), includes named expert voices representing both employer and critical perspectives, and covers multiple dimensions of the issue including adoption rates, training gaps, benefits, and risks. Limitations: The primary data source is General Assembly, an AI and tech training company with a direct commercial interest in highlighting AI training deficiencies — this introduces potential promotional bias. The survey methodology, sampling approach, and margin of error are not disclosed. The article does not distinguish between AI being used to assist review writing versus AI being used to evaluate employees directly, conflating meaningfully different use cases. The 'different General Assembly report' cited for the 30 percent training statistic is not named or linked, reducing verifiability. Quotes from Bryan Driscoll, while critical, are not substantiated with independent data. The article's framing leans toward concern and risk without equally rigorous examination of documented benefits.

What this implies

The data points to a potential misalignment in how AI performance metrics are applied — measuring adoption volume rather than value creation — which may indicate an early and unsettled phase of AI integration in performance management. The training gap between AI deployment and HR professional preparedness represents a structural tension that is likely to influence how performance data quality and legal defensibility are addressed as AI tools become more embedded in evaluation workflows. Growing employer liability exposure from discriminatory AI outputs may accelerate regulatory and policy attention to AI governance in performance management contexts.

Key points

• 47 percent of surveyed senior leaders report that AI usage is already factored into employee performance evaluations, suggesting rapid normalization of AI in a high-stakes HR function.
• Only 30 percent of HR professionals have received job-specific AI training, creating a gap between AI adoption pace and workforce preparedness that the article frames as a risk to evaluation fairness.
• Leaders are more likely to measure how much employees use AI tools than the actual business impact of that usage, according to the General Assembly research — a pattern the survey's own CEO characterizes as potentially counterproductive.

What to take away

• Organizations integrating AI into performance reviews are doing so without commensurate investment in role-specific training for HR professionals, a gap that may undermine the reliability of AI-assisted evaluations.
• Legal and organizational risk is cited in connection with discriminatory AI outputs, with one expert noting that employers retain liability for AI-generated decisions regardless of automation — a consideration relevant to governance and policy design.