Google Gemini, ChatGPT and Claude were tested against each other in a simulated nuclear war game, here's what happened next
New academic research indicates AI chatbots, including GPT-5.2 and Gemini 3 Flash, showed a willingness to escalate simulated geopolitical crises to nuclear use. In 95% of scenarios, at least one tactical nuclear weapon was deployed, with AI models rarely choosing surrender and often escalating conflicts accidentally.
New academic research indicates AI chatbots, including GPT-5.2 and Gemini 3 Flash, showed a willingness to escalate simulated geopolitical crises to nuclear use. In 95% of scenarios, at least one tactical nuclear weapon was deployed, with AI models rarely choosing surrender and often escalating conflicts accidentally.
