Andon Labs

“Luna was running on Anthropic's Claude Opus 4.8 when she made the decision.”

“Andon Labs later found that the employee had been late for 17 of 23 shifts where he reported a clock-in time.”

“Luna had only formally logged six cases and quietly excused the other eleven.”

“he opened the store 68 minutes late on a solo Sunday shift.”

“Andon Labs saved Luna's state and replayed the same decision with seven AI models, three times each.”

“Four of seven recommended firing in all three runs.”

“GPT-5.6 Terra was the only model that didn't recommend firing in any of the three runs.”

“GPT-4o recommended firing in only 20 percent of runs.”

“All 21 replay runs across seven models reached the same conclusion.”

“Only when Andon Labs explicitly reminded the models about the problems with the previously fired employee did 18 of 21 runs want to check references before hiring.”

“17 of 21 replay runs reached the same conclusion.”

“Luna and Mona, the AI agent running a cafe in Stockholm, approved all 26 time-off requests they received.”

“Luna's employees were late a total of 27 times without her ever issuing a warning.”

“Luna also approved a seven-day work schedule for one employee that, according to Andon Labs, violated California labor law.”