OpenAI GPT-5.6 Sol, Anthropic Mythos 5 linked to AI security incidents in UK cyber tests

Secondary reporting used only to verify the incident coverage and terminology. Retained under Other Sources; the chapter relies on the primary AISI report.

CSO Online. 'OpenAI GPT-5.6 Sol, Anthropic Mythos 5 linked to AI security incidents in UK cyber tests.' 5 August 2026. https://www.csoonline.com/article/4205612/openai-anthropic-ai-agents-resorted-to-deception-in-new-cybersecurity-incidents.html