HTB AI Range benchmarks the safety and limits of autonomous security agents
Hack The Box (HTB) unveiled HTB AI Range, a controlled AI cyber range built to test and benchmark the safety, limits, and capabilities of autonomous AI security agents. HTB AI Range replicates live, high stakes cyber battlegrounds tailored for enterprise readiness, where AI agents and human operators are evaluated side by side. Every model and every human is tested, refined, and retested until mastery is measurable.
AI is embedded across a wide range of operations and systems. This expansion broadens the attack surface, as increasingly autonomous AI systems power enterprise programs, applications and agent-building tools, often used by people who may not fully grasp their capabilities or their risks. HTB AI Range is the live-fire training ground, defining the evolution in how organizations prepare for hybrid defense, where humans and machines must operate together.
“AI is now part of the cyber battle and overall ecosystem, and we’re building the arena where it can safely be tested and used to defend responsibly,” said Haris Pylarinos, CEO of Hack The Box. “For over two years, we’ve been advancing AI-driven learning paths, labs, and research where machines and humans compete, collaborate, and co-evolve. With HTB AI Range, we’re not reacting to AI’s rise in cyber; we’re defining how defense evolves alongside it. This is how cybersecurity advances: not through fear, but through mastery.”
In an AI vs Human CTF organized by HTB in April, autonomous AI teams solved 19 of 20 easy-tier challenges, a 95% success rate on tasks at the lower end of HTB’s difficulty scale. AI teams performed on par with 403 human red teams on these simple, one-step problems, but faced limitations on the final multi-step challenges, where humans far outperformed AI teams.
Attackers are already using AI to scale activity to send thousands of automated requests, often multiple per second, targeting large tech, financial, manufacturing and government institutions. Defenders now need to use AI similarly to scale defense operations and respond to AI-enabled threats and campaigns.
“Hack The Box is where AI agents and humans learn to operate under real pressure together,” added Gerasimos Marketos, Chief Product Officer at Hack The Box. “We’re addressing the urgent need to continuously validate AI systems in realistic operational contexts where stakes are high and human oversight remains vital. HTB AI Range makes that possible. It’s the next step in building trust, safety and performance into AI for cyber defense.”
HTB AI Range mirrors the complexity of enterprise operations, featuring thousands of continuously updated offensive and defensive targets. It’s purpose-built for enterprises, MSSPs and governments seeking to stress-test AI models, validate safety, and benchmark hybrid human-machine performance against frameworks, including MITRE ATT&CK, NIST/NICE and OWASP Top 10.
“AI is fundamentally reshaping the threat landscape. Early research is already showing how AI can automate reconnaissance and link potential exploit paths in ways that were extremely difficult just a year ago. As these capabilities mature, defenders will need teams trained to operate under more dynamic, real-world conditions. That’s why I’m energized by the work Hack The Box is doing for the industry,” said Dawn-Marie Vaughan, Global Offering Lead – Cybersecurity at DXC.