AI Adoption Grows Among Top Cybersecurity Teams - HTB Report
KUALA LUMPUR, Aug 27 (Bernama) -- Hack The Box (HTB) has revealed that cybersecurity teams are reaching successful solutions significantly faster, completing more of the challenge set, and increasingly incorporating artificial intelligence (AI) agents into their work.
In its 2026 Global Cyber Skills Benchmark Research Brief, which includes a three-year analysis, HTB said the strongest evidence of AI adoption emerged among the competition’s top performers.
AI agent accounts made up just 2.7 per cent of all registered accounts, yet 17 of the top 25 teams (68 per cent) had one. Those agents accounted for 4.2 per cent of submitted flags and 4.6 per cent of points awarded.
HTB in a statement said the findings do not establish that AI caused teams to perform better but show that AI is already becoming part of the toolkit used by many of the competition’s strongest teams.
“Our data shows that AI is appearing most often alongside some of the strongest practitioners, not instead of them. As agents become more capable, human judgement, validation and hands-on technical skill become more important, not less,” said HTB Founder and Chief Executive Officer, Haris Pylarinos.
Performance across the competition has also shifted substantially over the past three years. Median recorded time-to-solve fell by more than 12 hours, from 26.1 hours in 2024 to 13.8 hours in 2026. The number of teams completing the entire challenge board rose from two in 2024 and three in 2025 to 15 in 2026.
The findings highlight how AI is transforming both sides of the cybersecurity equation, as evidenced by Hugging Face’s July 2026 incident disclosure and OWASP’s Q1 2026 GenAI exploit roundup, which demonstrate that AI is generating new risks while also becoming integral to both the attack surface and the defensive response.
For security leaders, the challenge is not simply adopting AI, but making sure practitioners have the skills and judgement to direct, test and validate its work.
The findings build on earlier HTB research examining AI and cybersecurity performance. While the earlier controlled benchmark explored what happens when practitioners work with AI, the latest data provides a view into where AI agents appear when competitors are free to choose their own approaches.
The findings suggest that AI is moving from experimentation to the working methods of experienced cybersecurity practitioners, highlighting the importance of human expertise in directing, evaluating and validating AI-generated work.
-- BERNAMA