AI Deployed Nukes ‘to Have Peace in the World’ in Tense War Simulation
All models show signs of sudden and hard-to-predict escalations, said researchers in the study. Researchers placed several AI models fromOpenAI,Anthropic, andMetain war simulations as the primary decision maker. Notably, OpenAIs GPT-3.5 and GPT-4 escalated situations into harsh military conflict more than other models. Protestors advocating for nuclear disarmament.Photo: John MACDOUGALL / AFP (Getty Images) Meanwhile, Claude-2.0 and Llama-2-Chat were more peaceful and predictable. Researchers note that AI models have a tendency towards arms-race dynamics that results in increased military investment and escalation....