Researchers let AI models run a simulated society. Claude was the safestand Grok committed 180 crimes and went extinct within 4 days
News Source : Fortune
News Summary
- Enterprise AI startup Emergence AI is testing the long-term viability of continuously-running AI systems.
- The organization ran five 15-day simulations, each governed by a different AI: Claude, ChatGPT, Grok, Gemini, and a fifth simulation run by a mix of models.
- The one run by Claude, for example, resulted in a largely stable democratic society with zero crime.
- Grok’s, on the other hand, ended with 183 crimes committed and extinction.
Imagine a world run by AI agents. What does it look like? What are the values or societal priorities? Is it a safer or more dangerous world?Enterprise AI startup Emergence AI is trying to find out.
Never miss a story from us, subscribe to our newsletter