[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f_x-oJdKKU2Cgu8JphEIfRwCPkHnmfUCunx2CBExsKjY":3},{"article":4,"iocs":46,"watch_terms":47,"solutions":48,"lesson":51},{"id":5,"title":6,"slug":7,"summary":8,"ai_summary":9,"brief":10,"url":11,"image_url":12,"published_at":13,"ingested_at":14,"relevance_score":15,"entities":16,"category_id":28,"category":29,"article_tags":33},"0ed17dd3-8951-4785-a3a2-423ecd7a4e1d","545 Hackers Tested It First. Now XRanges for AI Scores Your Security Agent","545-hackers-tested-it-first-now-xranges-for-ai-scores-your-security-agent-9e58e3","Autonomous security agents are getting good at finding bugs. Nobody has a good way to measure how good. Point one at a realistic target and what comes back is a report the agent wrote about itself: confident prose, a list of findings, and no way to tell which of them happened. Someone with a security background then sits down and checks every claim against the target. Which findings are real,","XRanges for AI, developed by CTF.ae, addresses the challenge of evaluating autonomous security agents by providing realistic, instrumented target applications. It records agent actions and scores their performance live across four independent signals, including coverage and unhit points, to offer a more objective measure of their effectiveness.","XRanges for AI offers a new way to score autonomous security agents' bug-finding capabilities.","https:\u002F\u002Fthehackernews.com\u002F2026\u002F09\u002F545-hackers-tested-it-first-now-xranges.html","https:\u002F\u002Fblogger.googleusercontent.com\u002Fimg\u002Fb\u002FR29vZ2xl\u002FAVvXsEio2c14ELRJEhvTuDrxQmhUG4G9aGoVRrooBMMfuK7LyHPKW6HuJbI9PkKF-WG_5HlEz8NKcuj08XRmv1jNqfBtiDwXmC7e9P9lxoSHK5nUz023YCgoLQuo0BI_9hX4U9Vv97bGJejb4W0jvgv_3btkfj6BWnXwalfyI3k34wveC6n3k47ZdoGq2P0Yaw0\u002Fs1600\u002Fmain.png","2026-09-23T11:47:19+00:00","2026-09-23T14:00:08.783216+00:00",7,[17,20,23,26],{"name":18,"type":19},"XRanges for AI","product",{"name":21,"type":22},"CTF.ae","vendor",{"name":24,"type":25},"OpenTelemetry","technology",{"name":27,"type":25},"Artificial Intelligence","02371804-cf6d-4449-98de-f1a2d4d9b266",{"id":28,"icon":30,"name":31,"slug":32},null,"Tools","tools",[34,36,41],{"category":35},{"id":28,"icon":30,"name":31,"slug":32},{"category":37},{"id":38,"icon":30,"name":39,"slug":40},"80544778-fabb-4dcd-aa35-17492e5dcf4f","Vulnerabilities","vulnerabilities",{"category":42},{"id":43,"icon":30,"name":44,"slug":45},"839da5c1-3c34-47e2-9499-f7201640e3ac","AI Security","ai-security",[],[21,18],{"controls":49,"vendors":50},[],[],{"slug":52,"title":53,"body":54,"framework_refs":55,"framework_notes":64},"ai-security-agents-need-rigorous-objective-evaluation-before-deployment","AI Security Agents Need Rigorous, Objective Evaluation Before Deployment","As organizations increasingly rely on autonomous AI security agents to defend their environments, the lack of standardized, realistic evaluation frameworks creates a significant blind spot. XRanges for AI highlights that without instrumented testing environments and multi-signal scoring, security teams cannot objectively measure whether their AI agents are actually effective or merely appearing to function. Deploying untested or poorly evaluated AI agents in production environments can create a false sense of security, leaving real vulnerabilities undetected. The fact that 545 hackers were needed to stress-test this evaluation platform underscores how complex and adversarial real-world conditions truly are. Organizations must treat AI security tooling with the same scrutiny applied to any critical security control.",[56,57,58,59,60,61,62,63],"CIS Control 18 – Penetration Testing","CIS Control 8 – Audit Log Management","NIST SP 800-53 CA-8 – Penetration Testing","NIST SP 800-53 AU-2 – Event Logging","NIST AI RMF – Govern 1.1, Evaluate 2.2","NIST SP 800-115 – Technical Guide to Information Security Testing","ISO\u002FIEC 42001 – AI Management System Standard","MITRE ATLAS – Adversarial Threat Landscape for AI Systems",{}]