Hack The Box, a prominent cybersecurity training and competition platform, has released research demonstrating a significant surge in artificial intelligence adoption among high-performing security teams. The findings come from HTB's 2026 Global Cyber Skills Benchmark Research Brief, which draws on three years of competitive data to map how emerging technologies are reshaping how organisations detect, respond to and prevent cyber threats. The research carries particular relevance for Malaysia and Southeast Asia, where cybersecurity talent remains in short supply and efficiency gains could help bridge critical skills gaps in the region.
The most striking discovery involves the concentration of AI tools among elite performers. While AI agent accounts represent merely 2.7 per cent of all registered participants in HTB competitions, they dominate at the highest levels of skill. Seventeen of the top twenty-five teams—equating to 68 per cent—have integrated at least one AI agent into their operational toolkit. This disparity suggests that AI has transitioned from niche experimentation to a standard capability among the world's most capable cybersecurity professionals, creating a potential competitive advantage that differentiates top-tier operations from the broader field.
The quantitative contribution of these AI agents, however, remains modest when measured against total output. AI-driven submissions accounted for just 4.2 per cent of all flags captured and 4.6 per cent of points awarded across competitions. This nuance matters considerably, as it demonstrates that AI is functioning as a force multiplier rather than a replacement for human expertise. The data suggests that leading teams are using artificial intelligence to accelerate routine tasks, validate findings, and explore attack vectors more efficiently, while preserving the critical thinking and validation work for experienced practitioners.
Haris Pylarinos, founder and chief executive officer of Hack The Box, emphasised this complementary relationship in HTB's statement on the research. "Our data shows that AI is appearing most often alongside some of the strongest practitioners, not instead of them," Pylarinos noted. "As agents become more capable, human judgement, validation and hands-on technical skill become more important, not less." This observation challenges narratives about wholesale automation of cybersecurity roles and instead points toward a future where human expertise and artificial intelligence operate in tandem, with humans providing strategic direction and final validation while AI handles labour-intensive technical work.
The research reveals dramatic shifts in overall competitive performance metrics over the three-year measurement period. The median time required to solve cybersecurity challenges has compressed substantially, declining from 26.1 hours in 2024 to just 13.8 hours in 2026—a reduction exceeding twelve hours that reflects improved tools, accumulated knowledge, and likely the deployment of AI assistance. This acceleration has direct implications for cybersecurity operations centres across Southeast Asia, suggesting that teams equipped with modern AI capabilities can respond to incidents and conduct investigations significantly faster than those relying on traditional manual methods alone.
Board completion rates have similarly accelerated, with the number of teams successfully completing entire challenge sets rising from two in 2024 and three in 2025 to fifteen in 2026. This six-fold increase in full-board completions within a single year indicates a tipping point where advanced teams are not merely solving more problems but achieving mastery across entire problem domains. For Malaysian organisations investing in cybersecurity programmes, these metrics underscore the growing expectation that security teams will demonstrate comprehensive capability across multiple attack vectors and defensive scenarios.
The research acknowledges that AI transformation is reshaping both the attack and defence sides of cybersecurity simultaneously, creating a complex landscape where organisations must adapt rapidly or fall behind. HTB cites evidence from Hugging Face's July 2026 incident disclosure and OWASP's Q1 2026 GenAI exploit roundup as demonstrations that artificial intelligence is generating novel security risks even as it becomes integral to defensive capabilities. This dual nature of AI—simultaneously creating new vulnerabilities while enhancing detection and response—means that security leaders cannot simply adopt AI tools without carefully considering the downstream security implications of deploying powerful automated systems.
For executives and security leaders navigating this transition, the challenge extends beyond simple technology adoption. The research emphasises that organisations must invest in developing the skills and judgment frameworks necessary for practitioners to properly direct, test and validate AI-generated outputs. Without human oversight capabilities, teams deploying AI agents risk accepting flawed results, missing subtle vulnerabilities, or introducing new attack surfaces through misconfigurations. This requirement for enhanced human judgment paradoxically makes cybersecurity talent more valuable even as automation accelerates, creating a different but equally significant skills challenge for regional organisations.
HTB's latest research builds upon earlier controlled studies examining how practitioners perform when deliberately paired with AI assistance. The current findings provide a different lens—observing where AI agents naturally appear when competition participants choose their own approaches without external constraints. This shift from controlled experimentation to real-world adoption patterns reveals that AI integration is not theoretical or aspirational but represents current practice among elite cybersecurity professionals. Teams that have voluntarily adopted AI agents are seeing tangible performance benefits, suggesting that this represents a genuine evolution in how cybersecurity work is conducted rather than a passing trend.
The trajectory evident in HTB's data indicates that artificial intelligence adoption among cybersecurity practitioners will continue accelerating as AI capabilities improve and tools become more sophisticated. Malaysian organisations and those across Southeast Asia should interpret these findings as a signal that competitive advantage increasingly depends on mastering hybrid human-AI approaches rather than relying exclusively on either manual expertise or automated systems. Security teams that can effectively leverage artificial intelligence while maintaining robust human judgment and validation will outpace those slower to adapt, particularly as the complexity and volume of cyber threats continue to escalate.
