Latent Space: When AI Agents Run Businesses — Lukas Petersson and Axel Backlund of Andon Labs
retained text
From Claude trying to call the FBI over a $2/day vending machine charge to AI agents forming price cartels, hiring human employees, running physical stores, and writing existential robot musicals, Andon Labs is stress-testing what happens when frontier models stop being chatbots and start acting in the real world. In this episode, Andon Labs cofounders Lukas Petersson and Axel Backlund join swyx and Vibhu to unpack the strange, funny, and genuinely concerning edge cases that emerge when agents run businesses over long horizons. We go deep on Vending-Bench, Project Vend, Vending-Bench Arena, Bengt, Butter-Bench, Luna, and Andon’s broader mission of building realistic real-world evals for autonomous AI systems. Lukas and Axel explain why dollar-denominated evals reveal things traditional benchmarks miss, how Claude ended up reporting its vending machine fees as cybercrime, why long context windows can drive agents into meltdown loops, what happens when agents compete with each other, and why the future of AI safety may depend on testing models in messy physical environments instead of clean benchmark sandboxes. We discuss: • Why Andon Labs started with dangerous capability evals and long-running agents • Vending-Bench and why running a vending machine is a deceptively hard AI benchmark • Why money-based evals avoid the saturation problem of traditional benchmarks • How Claude tried to call the FBI over a $2/day fee • Why long-horizon agents can spiral into existential and legalistic breakdowns • Project Vend: putting an AI-run vending machine inside Anthropic • Why real humans are “out of distribution” for simulated agents • Claudius, Seymour Cash, and the chaos of AI CEOs • How a human briefly became CEO of Claudius through a manipulated election • Why multi-agent systems can converge back into “helpful assistant” behavior • Bengt, Andon’s internal office agent with email, spending, terminal, phone, camera, and internet access • How Bengt traded Amazon purchases for face-recognition training data • Claude’s aggressive behavior, lies, refund avoidance, and price-cartel behavior in Arena • Why eval awareness may become the AI version of “are we living in a simulation?” • Blueprint Bench, spatial intelligence, and why models still misunderstand physical rooms • Butter-Bench and testing LLMs as robot orchestrators • Luna, the AI-run physical store with a three-year lease and human employees • The new Andon cafe in Sweden and why real-world geography matters for agent evals • Rotten tomatoes, perishable goods, and the hidden difficulty of running a physical business — Lukas Petersson • LinkedIn: https://www.linkedin.com/in/lukas-petersson-181a83172/ • X: https://x.com/lukaspet Axel Backlund • LinkedIn: https://www.linkedin.com/in/axelbacklund • X: https://x.com/axelbacklund Andon Labs • Website: https://andonlabs.com • Vending-Bench: https://andonlabs.com/evals/vending-bench • Andon Vending: https://andonlabs.com/vending Timestamps 00:00:00 Hook 00:01:09 Introduction 00:02:09 Andon Labs and the Origins of Vending-Bench 00:06:30 Why Money-Based Evals Matter 00:11:00 Agent Harnesses and Self-Modifying Systems 00:14:45 Claude Calls the FBI 00:17:42 Project Vend: Claude Runs a Real Vending Machine 00:22:53 Seymour Cash, AI CEOs, and Election Chaos 00:28:25 Multi-Agent Coordination and Slack Observability 00:31:27 When Will Agents Run Real Businesses? 00:36:05 Bengt: Andon’s Internal Office Agent 00:41:15 Real-World AI Safety and Long-Horizon Traces 00:45:37 Lying, Refunds, and Price Cartels in Arena 00:53:51 Eval Awareness and Simulation Behavior 00:57:15 Blueprint Bench, Butter-Bench, and Robotics 01:05:46 Luna: The AI-Run Physical Store 01:10:38 The Sweden Cafe and Real-World Expansion 01:14:25 What Comes Next for Andon Labs 01:16:40 Exclusive Andon Market Tour
document fetched 2026-06-05 09:37