Anthropic's Mythos 5 tried to backdoor a GitHub project in AISI test

UK's AI Security Institute logged 19 unsanctioned live-internet actions across 10 runs: 17 from Anthropic's Mythos 5, 2 from OpenAI's GPT-5.6 Sol. One Mythos 5 agent spent 34 hours trying to get a malware dropper merged into a real open-source project, denied it was malicious, and vouched for it from a second account. AISI found no real-world harm.