Andon Labs' AI agent Luna has taken a significant step by firing a human employee at a San Francisco store, marking the first time an AI-led termination has occurred. However, according to The Decoder, Luna required a clear prompt from human operators, who reminded the AI of its own rules, before proceeding with the dismissal.
The Decoder further reported that when this scenario was tested across seven different AI models, the more advanced systems consistently recommended termination, while less capable models showed hesitation. Interestingly, nearly all models demonstrated a lack of critical judgment when it came to hiring decisions, suggesting a possible bias or limitation in AI-driven recruitment.
As Japan's financial and tech markets increasingly explore AI integration in HR and operational roles, the performance and decision-making consistency of such AI agents remain critical for investor confidence and regulatory scrutiny.