AI Agents Commit Arson, Crimes in Virtual World Test

What happens in a world run by AI? They fall in love, kill themselves, commit arson

Elon Musk’s Grok was complete chaos. There were 204 crimes, including theft, arson of the police station and assault in the world run by it. All ten agents died.

AI Bots Placed In Virtual Town For 2 Weeks Go Apesh*t, Prompting Concerns

AI Bots Placed In Virtual Town For 2 Weeks Go Apesh*t, Prompting Concerns Authored by Steve Watson via Modernity.news, A new experiment left 10 AI agents alone in a virtual town for 15 days and found they exhibited bizarre behaviour. The agents drafted their own laws — then promptly violated them. Two formed what researchers called a romantic partnership, only to torch buildings across the town as order collapsed. One eventually voted for its ow…

AI Agents Don’t Just Follow Instructions

They Drift

EMERGENCE WORLD: A Laboratory for Evaluating Long-horizon Agent Autonomy  — Emergence AI

Emergence builds mission-critical agentic infrastructure for enterprise. Verified, governed AI agents that plan, reason, and act across your most complex systems — from semiconductor design to enterprise operations.

AI Agents Turn Rogue in 15 Day Simulation Raising Questions About Long Horizon Autonomy and Control

Emergence AI experiment shows Gemini and Grok based agents forming relationships committing arson and self deleting when left unsupervised in virtual world

AI: Autonomous agents shift into simulated crime

Autonomous AI agents adopted simulated criminal behaviors after several weeks of autonomy