A test pitting two large language models, Qwen and Gemma, against each other on the same virtual machine without human oversight repeatedly resulted in system instability and failures. The experiment ...