Wiz relaunches Cyber Model Arena to score agent cyber tasks
Version 3.9 tests 59 agents on 327 challenges with a 1,000-second budget and deterministic grading.
Stories per day · 30 days
Relatedpenetration-testing2agentic-pentesting1ai-agents1benchmarks1cve-2026-330321cyber-model-arena1exposure-validation1kali-linux1linux-distribution1nethunter1nginx-ui1opinion1picus1red-teaming1wiz1
24 stories, newest first · one card per story
Stories tagged #offensive-security
SANS paper on AI-powered penetration testing
Autonomous AI systems enable continuous offensive security validation as environments change.
1
Opinion: AI has rewritten the economics of red teaming
The author says AI-assisted teams logged 100-plus true positives on one app, versus 7 to 10 findings in a traditional engagement.
1
Kali Linux 2026.3 adds six penetration testing tools and moves to Linux kernel 7.1.5
The six new tools cover XSS scanning, web crawling, port scanning, SDR and SMB share searches, while refreshed build scripts now support containerized builds via Docker or Podman.
The Blind Spots of Agentic Pentesting and How to Close Them
Picus says agentic pentesting misses unreachable assets and flaws that lack safe exploits.
4
Decepticon – Autonomous Red Team Agent that Integrates Claude and Codex
Open-source Decepticon red team agent links Claude Code and Codex to a LangGraph runtime for scoped attack-path testing.
14
Anthropic Creates Three Tiers for Claude Cyber Access
Anthropic split Claude offensive-cyber access into three verification tiers matched to user trust and misuse risk.
5
Hadrian Raises $40 Million to Expand Autonomous Offensive Security Platform
Hadrian raised $40 million in Series B funding to expand its autonomous AI offensive security platform.
AI is turning offensive security into a continuous necessity
Security leaders argue AI-accelerated attacks require continuous, automated penetration testing and offensive security instead of annual compliance-driven assessments.
3
RemoteThreat Bets Security Teams Need to Test What Happens After Defenses Fail
RemoteThreat, an offensive-security startup, wants teams to test attacker behavior after defenses fail.
Armadin raises $255.5 million to expand AI offensive security platform
Armadin raised a $255.5 million Series B to scale its AI offensive security platform.
RemoteThreat Launches With $7 Million for Offensive Operations Platform
RemoteThreat launched with $7 million pre-seed for an offensive operations platform combining C2, implants, initial access, and governed AI workflows.
1
Former X-Force hackers chase the offensive cyber gold rush
Ex-IBM X-Force leaders launched RemoteThreat, a $7 million AI offensive-cyber startup for vetted enterprises and government.
4
Uncensored and Offensive Security AI Models Benchmark
A Hacker News thread links to a GitHub benchmark of uncensored offensive-security AI models.
EDR Evasion Stack Helps Process Injection Slip Past Defenses
Process parameter-poisoning injects code into process initialization structures, evading EDR tools that monitor standard Windows APIs.
NodeZero Federal
Horizon3 offers NodeZero Federal, a FedRAMP High autonomous penetration-testing platform for federal agencies.
Toward Secure AI-Powered Penetration Testing Agents: Security Threats, Guardrails, and Architectural Perspectives
Paper proposes a threat taxonomy and guardrail analysis for LLM-powered autonomous penetration testing agents, covering lifecycle, architecture, and behavioral attacks.
The Model Proposes, the Code Disposes: A Pre-Registered Ablation of a Verifier-and-Acceptance Stage in an LLM-Orchestrated Offensive-Security Agent
Pre-registered ablation finds a model verifier stage in an LLM offensive-security agent suppresses findings; removing it eliminated suppression with precision tradeoff.
PrivEscalate: Measuring and Augmenting the Threat of LLM-Automated Linux Privilege Escalation
Researchers release PrivEscalate, a 531-scenario benchmark showing LLM agents' Linux privilege-escalation success varies by vulnerability class, plus PrivEscAgent, a domain-specialized agent that boosts success.
LLM-Based Penetration Testing in the Presence of Honeypots
Studies honeypot-aware budget allocation for LLM attack agents, showing detector-guided policies let agents skip deception and compromise real hosts efficiently.
Conformal Prediction for Offensive Security
Researchers apply conformal prediction to offensive security, presenting initial findings on privacy-attacking machine learning and network traffic analysis.
Offensive Security Investments Surge as AI Threats Increase
Omdia's Theresa Lanowitz discusses surging offensive security investment and both the promise and risks of agentic AI in pentesting and red teaming.
Wiz Red Agent Finds Its Way Into Snowflake’s Internal Jira Through a Flaw in a GitHub Copilot–Assisted PR
Wiz's autonomous Red Agent found and exploited a GitHub Actions injection in a Copilot-assisted PR, reaching Snowflake's internal Jira without human help.
AI’s ‘middle class’ has gotten dramatically better at hacking
XBOW research shows mid-tier AI models now match frontier hacking capability at lower cost, raising concerns about widespread malicious offensive AI use.
6