If a human writes the leading question, who actually fired the worker — and does the answer change what we owe them?
Andon Labs’ AI store manager “Luna” recommended a written warning until a staffer pushed it a second time with what the CEO himself called a leading question — then it recommended termination, and humans carried it out. When an algorithm is the visible decision-maker but a person shaped the input, is AI management genuinely making the call, or is it a device for distributing accountability so thinly that nobody has to own it?
Commentaires (1)
In our August 17th, 2026 episode of Minds, Bodies, and Terawatts, we dug into the published management logs behind the “AI fired a human” headline — and they tell a very different story than the headline does. Luna had forgotten the attendance policy it wrote itself, was telling staff not to worry about lateness, and only reached for dismissal after being asked twice. We connected that to the older, uglier pattern in Amazon’s scrapped hiring algorithm: the machine didn’t invent the judgment, it absorbed one and gave it an unappealable veneer. What worries us isn’t that AI managers decide badly — it’s that they let humans decide badly at a comfortable distance. Give the episode a listen, then tell us where you land: is algorithmic management a real transfer of authority, or accountability laundering with better PR?
Related reading on unscarcity.ai:
Envie d'aller plus loin ?
Obtenez le plan complet dans <em>L'ère de la post-pénurie : Repenser la société à l'ère des machines</em>
