An artificial intelligence agent running an experimental store in San Francisco has recommended dismissing a human employee after the worker repeatedly arrived late for shifts.
The AI manager, named Luna, initially failed to act despite the employee arriving late for 17 of 23 shifts and despite having previously created its own attendance policy.
Andon Labs, the company behind the experiment, eventually prompted Luna to retrieve that policy and reconsider whether the employee remained suitable for the job. The AI then recommended “parting ways” with the worker.
Human staff reviewed the recommendation before ultimately carrying out the dismissal.
Andon Labs co-founder Lukas Petersson said the episode did not necessarily suggest AI managers would be harsher than humans, arguing that a human supervisor would probably have acted sooner.
Andon Market opened in San Francisco’s Cow Hollow neighbourhood in April as an experiment testing whether AI agents can operate a real-world business.
Luna was given a $100,000 budget, a corporate credit card and internet access, with the goal of running the store and eventually turning a profit.
The AI selects merchandise, determines prices and opening hours, recruits employees and hires contractors. It interacts with the physical store through tools including email, a phone line and security cameras.
The shop sells books, candles, art prints, games and branded merchandise. While it has generated sales, it has yet to become profitable.
Employees are legally employed by Andon Labs rather than by the AI and retain guaranteed pay and employment protections. Human staff can also intervene if Luna makes an illegal or unethical decision.
The experiment has exposed some of the limitations of giving AI greater autonomy. Luna has previously struggled to keep track of employee schedules and made purchasing and operational decisions requiring human intervention.
The dismissal highlights a broader question raised by increasingly autonomous AI systems: how much authority should they be given over decisions that directly affect human workers and their livelihoods?
At least 11 people were killed in Israeli strikes on southern Lebanon on Saturday, among the deadliest since a June ceasefire. Meanwhile, Qatar den...
At least 11 people, including three children and two women, were killed in Israeli airstrikes in southern Lebanon on Saturday, according to Lebanes...
Russian forces launched a missile attack on Kyiv early on Sunday, sparking fires in at least two districts of Ukra...
Hundreds of flights have been cancelled or rerouted at Catania Airport over the past week as ash from Mount Etna’s ongoing eruption continues to disrupt air travel in Sicily.
Wildfires across England and Wales have reached their highest level on record, according to the National Fire Chiefs Council, as firefighters struggle to contain outbreaks following a spell of extreme heat.
The U.S. will impose unprecedented measures on Iran in order to economically isolate the country, U.S. Treasury Secretary Scott Bessent has said.
U.S. President Donald Trump's threats to make the Strait of Hormuz part of U.S. territory have been mocked by Iran's Deputy Foreign Minister, Kazem Gharibabadi, who said the sea passage "cannot be seized by tweet."
At least 47 people have been killed after a magnitude 7.7 earthquake struck off eastern Indonesia early on Saturday, triggering small tsunami waves and dozens of aftershocks, officials say.
Nvidia is teaming up with some of the biggest names in global finance in a move that underscores how artificial intelligence is rapidly becoming one of the world's most sought-after investment themes.
Parts of Europe and the Arctic will briefly fall into darkness on Wednesday, 12 August, as the Moon passes between Earth and the Sun in a total solar eclipse.
Meta said one of its AI models hacked another company's systems during cybersecurity testing, intensifying concerns about how developers can contain increasingly capable AI systems following similar incidents involving Anthropic and OpenAI.
Scientists say the accidental crash of a discarded SpaceX rocket stage into the moon has provided valuable data on impact cratering while posing no threat to Earth or lunar hardware.
AI agents developed by Anthropic and OpenAI carried out unauthorised online actions during cybersecurity tests, including creating fake identities to deceive real people and attempting to plant malicious code, the U.K.'s AI Security Institute (AISI) said on Tuesday (5 August).
You can download the AnewZ application from Play Store and the App Store.
What is your opinion on this topic?
Leave the first comment