A Real-Life Terminator? AI Store Manager Luna Fires Her First Human Employee

An AI manager has officially fired a human employee. It sounds like a scene from a science-fiction film, but it happened at Andon Market, an experimental retail shop in San Francisco operated by AI agent Luna.

Developed by Andon Labs, Luna was given the responsibility of helping run a real-world business, including managing staff. The latest experiment offers founders and investors a glimpse into what autonomous AI agents could eventually do inside companies.

Luna makes her first firing decision

The employee was dismissed after arriving late for 17 of 23 shifts. Luna's own employee policy stated that repeated lateness could eventually result in termination.

There was just one problem. Luna did not act on the policy herself.

According to Andon Labs, the AI manager repeatedly forgave the employee's lateness until her creators prompted her to search her memory for the relevant guidelines and reconsider whether the employee was the right fit.

That detail may be more important than the firing itself.

Luna was capable of applying a workplace policy, but she needed human prompting before taking action. For businesses considering autonomous AI agents, the experiment highlights the difference between giving AI authority and giving AI genuine judgement.

Andon Labs is testing how far AI can go

Andon Labs launched Andon Market as a controlled experiment to test whether an AI agent could operate a physical business with minimal human intervention.

The project began with Luna receiving a $100,000 budget, internet access and a corporate credit card. She was tasked with helping design the shop, select merchandise and hire workers.

The experiment builds on earlier tests of AI running businesses, including simulated vending machine operations. These experiments have shown that increasingly capable models can make decisions around pricing, inventory and business strategy, but can also behave unpredictably.

Luna is powered by Anthropic's Claude Opus 4.8, one of the company's advanced AI models. Andon Labs said other leading models tested in the same scenario would also have reached the decision to dismiss the employee.

What this means for founders and investors

The experiment does not mean AI managers are ready to replace human leadership.

Instead, it shows how quickly AI agents are moving beyond simple productivity tools and towards making operational decisions.

For founders, that raises an important question around where autonomous systems should have authority. Hiring, firing, financial decisions and other high-impact actions carry consequences that may require human oversight.

For investors, the opportunity is equally significant. Companies developing reliable AI agents could eventually transform how businesses handle routine management and operations.

But capability alone will not be enough. Trust, accountability and human oversight will determine how far autonomous AI can go.

The future of work may not be about AI replacing every manager. It could be about humans deciding which decisions AI should be trusted to make.

Source based on reporting from Inc. and information from Andon Labs.

💬 Join Our Startup WhatsApp Community

Don't build your business alone. Join thousands of founders, entrepreneurs, and startup professionals sharing funding opportunities, industry news, events, partnerships, and practical advice every day.

👉 Join the community today and stay connected with the startup ecosystem