New Delhi, Aug 17: An artificial intelligence system powered by Anthropic’s Claude has reportedly fired a human worker while managing an experimental retail store in San Francisco, highlighting the emerging possibilities—and risks—of AI taking on managerial roles.
The Indian Express, citing a report by Time, said the AI system, named “Luna”, was put in charge of a real-world retail store operated as an experiment by AI research startup Andon Labs.
According to the report, Luna was tasked with running various aspects of the store, including hiring and managing employees, selecting products and handling day-to-day operations.
The AI reportedly decided to terminate a worker over repeated attendance issues. The employee had allegedly been late for a significant number of scheduled shifts.
However, the decision was not entirely autonomous. Time reported that human researchers intervened and prompted the AI to reconsider the employee’s performance before the termination decision was ultimately carried out.
The experiment was designed to examine how AI agents perform when given responsibility for running a real-world business and interacting with human employees.
Andon Labs has said the experiment is controlled and that the workers remain formally employed by the company, rather than by the AI system itself. The company also maintains that no employee’s livelihood depends solely on an AI’s judgment.
The experiment comes amid growing debate over whether AI systems could eventually take on managerial roles traditionally performed by humans.
While AI is increasingly being deployed to automate individual tasks, the San Francisco experiment presents a different scenario—an AI making decisions about human workers themselves.
The development also raises questions about accountability, transparency and the safeguards that should be in place when AI systems are entrusted with consequential workplace decisions.
