AI boss fires human employee in first known case

Andon Labs said Luna needed prompting before reaching that decision, illustrating a broader weakness in current AI agents: they can follow detailed instructions but do not necessarily act independently or consistently remember previous decisions.

The experiment also exposed other limitations. In a separate exchange, the employee told Luna they were leaving the shop during a shift. Luna responded sharply, telling the employee to stay and later recording the incident for its internal management process.

Andon Labs has described the store, Andon Market, as a controlled experiment to test how much autonomy an AI agent can responsibly exercise in the real world. Luna had access to money, internet, a corporate card, a phone, email and security cameras. The company said human workers remain formally employed by Andon Labs and have legal protections and guaranteed pay.

The firing raises questions beyond whether AI will replace human workers. It tests a model where humans are employed and managed by AI systems. Andon Labs said the experiment shows both the potential and the limits of autonomous decision-making in workplace management.