Microsoft launched MAI-Cyber-1-Flash, a compact cybersecurity model that scores 96 percent on the CyberGym benchmark when embedded in its MDASH multi-agent system, with Microsoft claiming costs should…
Wiz Research has built Atlas, an autonomous AI system for vulnerability research that ranks #1 on the CyberGym benchmark with a 90.9% success rate and has uncovered more than 200 previously unknown vu…
Microsoft's new MDASH (Microsoft Security multi-model agentic scanning harness) system, using over 100 specialized AI agents, discovered 16 previously unknown vulnerabilities in Windows networking and…
Microsoft introduced Project Perception, a new agentic security system designed for the age of AI, which coordinates red, blue, and green team agents to continuously perceive, reason, and act against …
Microsoft announced MAI-Cyber-1-Flash, a compact security model integrated into its MDASH multi-agent vulnerability harness, achieving 96% on the CyberGym benchmark, 12 points above Mythos, while deli…
Sakana AI released Fugu-Cyber, a cybersecurity-specialized orchestration model that reports 86.9% on CyberGym and 72.1% on CTI-REALM, surpassing reported scores for frontier models like GPT-5.5-Cyber …
A new benchmark called ExploitGym finds that frontier AI agents can autonomously turn real-world security vulnerabilities into working exploits, achieving unauthorized code execution on 869 tasks span…
Sakana AI released Fugu-Cyber, a new orchestration model for cybersecurity that achieves state-of-the-art performance with an 86.9% success rate on CyberGym and 72.1% on CTI-REALM, comparable to leadi…
The UK's AI Security Institute (AISI) found universal jailbreaks in OpenAI's GPT-5.6 Sol that allowed testers to bypass safeguards for high-risk cyber work, including vulnerability discovery and explo…
Google introduced Gemini 3.5 Flash Cyber, a lightweight cybersecurity model fine-tuned from its 3.5 Flash model to find, validate, and patch vulnerabilities quickly and efficiently. The model will be …
Microsoft's MDASH multi-agent AI system scored 88.45% on the CyberGym benchmark, surpassing Anthropic's Claude Mythos Preview (83.1%) and OpenAI's GPT-5.5 (81.8%) in vulnerability detection, and disco…
Microsoft's July 2026 security update included 1,150 CVEs, a threefold increase over the pre-AI baseline average of 383, according to an engineer who analyzed twelve months of Microsoft's Security Upd…
The National Academies of Sciences, Engineering, and Medicine published a report on AI for cybersecurity that cites Anthropic's Mythos model as capable of finding 83.1% of vulnerabilities in CyberGym,…
OpenAI released an improved GPT-5.5-Cyber model for vulnerability discovery and patching, expanded its cybersecurity partner program, and launched a Patch the Planet initiative to secure 30 open sourc…
Microsoft has introduced MDASH, a multi-model agentic security platform that uses over 100 specialized AI agents to automate large-scale vulnerability discovery across Windows, Hyper-V, and Azure code…