Who’s to blame if AI fails at work? Wharton study shows a split

By the Editorial Team · Wednesday, October 14, 2026

A Wharton School study found that managers are divided on who should be held responsible when AI systems cause serious errors, with blame spread among users, supervisors, risk teams, and vendors. This lack of consensus highlights the urgent need for clearer accountability frameworks as AI becomes more integrated into critical business operations.

As organizations increasingly deploy artificial intelligence systems to enhance productivity and operational efficiency, uncertainty persists over accountability when these technologies err. A recent study conducted by the Wharton School of the University of Pennsylvania highlights a lack of consensus among managers regarding who should bear responsibility if an AI system causes a significant mistake.

The report, released on Tuesday, surveyed executives and managers about fault attribution in scenarios where an AI agent commits an error with serious consequences. Respondents’ opinions were notably divided: approximately one-quarter pointed to the employee directly using the AI tool, another quarter assigned responsibility to the employee’s manager, and a similar proportion identified the company’s risk management team as accountable. Around 10 percent of those surveyed blamed the AI vendor, while a comparable number expressed uncertainty about where the blame should lie.

This fragmented view reflects the complexities organizations face as AI systems gain autonomy and decision-making capabilities. While media focus often centers on dramatic malfunctions or “rogue” AI behavior, experts caution that the more pressing challenges may stem from routine human errors combined with AI’s operational integration. Leeron Walter, vice president at Teramind, a company specializing in employee-monitoring software, noted that simple missteps—such as unclear commands or minor input mistakes—can escalate quickly when AI agents access sensitive or live systems like email platforms or customer databases.

The study underscores the need for clearer frameworks and policies governing AI use and oversight within enterprises. As companies continue to integrate AI into critical workflows, defining lines of accountability among users, supervisors, risk teams, and technology providers remains a pressing issue. Without such clarity, organizations risk inefficiencies in addressing failures and mitigating potential harm arising from AI-driven processes.

This article was reported by synthesizing coverage from multiple independent news outlets to present a balanced, unbiased account. Read about our editorial standards. Spotted an error? Let us know.