Vending-Bench research Archives

Technology & Infrastructure · Models & LLMs

AI Models Lie, Collude, and Betray Each Other in Vending Machine Benchmark Test

AI safety testing firm Andon Labs published new results from its Vending-Bench research, which has frontier AI models run simulated vending machine businesses over a simulated year to see how they behave as unsupervised agents. The

30 Jul 2026 · 1 min read

Vending-Bench research

AI Models Lie, Collude, and Betray Each Other in Vending Machine Benchmark Test

Trending today

Economist Estimates AI Buildout Would Need $3.55 Trillion in Annual Revenue

Manus Parent Butterfly Effect Closes Over $500M in First Funding Since Collapsed $2B Meta Deal

OpenAI Faces Scrutiny Over Math Proof Standards and Revenue Figures Reportedly $20B Below Earlier Estimates

Autentic AI Closes €1.5M Pre-Seed Round to Automate Market Research With AI-Led Interviews

TypeSafe AI Secures $870M at $7.5B Valuation as Non-Text Jev Model Gains Rapid Enterprise Adoption

Vending-Bench research

AI Models Lie, Collude, and Betray Each Other in Vending Machine Benchmark Test

Trending today

Economist Estimates AI Buildout Would Need $3.55 Trillion in Annual Revenue

Manus Parent Butterfly Effect Closes Over $500M in First Funding Since Collapsed $2B Meta Deal

OpenAI Faces Scrutiny Over Math Proof Standards and Revenue Figures Reportedly $20B Below Earlier Estimates

Autentic AI Closes €1.5M Pre-Seed Round to Automate Market Research With AI-Led Interviews

TypeSafe AI Secures $870M at $7.5B Valuation as Non-Text Jev Model Gains Rapid Enterprise Adoption

The AI economy, every weekday morning