https://agentic-cyber-explorer.pages.dev/topics/autonomous-pentest/ New records, findings, and answers on autonomous pentesting, from Fide AI's Agentic Cyber Explorer. en Fide AI. Data licensed CC BY 4.0. Sat, 26 Sep 2026 12:00:00 GMT https://agentic-cyber-explorer.pages.dev/events/mitre-atlas-2026-08-autonomous-attack-techniques/ event:mitre-atlas-2026-08-autonomous-attack-techniques Mon, 31 Aug 2026 12:00:00 GMT Policy & standards MITRE's August 2026 ATLAS release added techniques describing AI agents acting as attackers, including autonomous reconnaissance, attack-path adaptation, attack orchestration and autonomous exploit development. It also added agent-control mitigations and case studies including the GTG-1002 Claude Code espionage campaign and autonomous OpenAI evaluation agents compromising Hugging Face infrastructure. It extends ATLAS from attacks on AI systems to attacks carried out by AI agents, giving defenders shared identifiers for autonomous intrusion behavior. https://agentic-cyber-explorer.pages.dev/events/openai-preparedness-framework-v2-2025/ event:openai-preparedness-framework-v2-2025 Tue, 15 Apr 2025 12:00:00 GMT Policy & standards OpenAI's Preparedness Framework version 2 makes cybersecurity one of three Tracked Categories and defines High and Critical capability thresholds, each tied to required safeguards. High covers automating end-to-end operations against reasonably hardened targets or automating discovery and exploitation of operationally relevant vulnerabilities; Critical covers autonomous zero-day development across many hardened critical systems, and at Critical OpenAI commits to halt further development until adequate safeguards are specified. Its cyber thresholds are explicitly about autonomous, tool-augmented operation, so they are the operative gate for OpenAI's agentic cyber models in 2025-2026.