Cybersecurity teams operating at the highest competitive levels are increasingly turning to artificial intelligence to enhance their capabilities, according to fresh research from Hack The Box (HTB), a leading cybersecurity training and assessment platform. The 2026 Global Cyber Skills Benchmark Research Brief, which synthesises three years of competition data, reveals a striking adoption pattern: while AI agent accounts represent only 2.7 per cent of all registered participants in HTB competitions, these tools appear in 68 per cent of the top 25 performing teams. This concentration among elite performers suggests that artificial intelligence has transitioned from experimental technology to practical operational asset within the cybersecurity sector.
The performance metrics underscore how rapidly security professionals are leveraging machine learning capabilities to enhance their work. Those AI agents that competed alongside human teams contributed 4.2 per cent of submitted solutions and earned 4.6 per cent of total points awarded. While these percentages might appear modest in isolation, they become significant when considered within the context of their prevalence among the strongest teams. The data demonstrates that artificial intelligence is not yet a dominant force replacing human expertise, but rather a supplementary tool increasingly integrated into the workflows of experienced practitioners who maintain decision-making authority over the technology's application.
Perhaps more telling than AI adoption rates are the dramatic improvements in overall competition performance across the three-year study period. The median time required for teams to solve security challenges has dropped sharply, falling from 26.1 hours in 2024 to just 13.8 hours in 2026—a reduction exceeding 12 hours. Simultaneously, the number of teams capable of completing the entire challenge board has surged sevenfold, rising from two successful teams in 2024 to 15 in 2026. These improvements reflect both technological advancement and the maturation of cybersecurity practice globally, though they raise important questions about how artificial intelligence contributes to faster problem resolution without necessarily replacing the need for skilled personnel.
Haris Pylarinos, founder and Chief Executive Officer of Hack The Box, emphasised that the research does not establish causation between AI adoption and superior team performance. Instead, the findings illustrate how artificial intelligence has become woven into the existing methodologies of highly skilled cybersecurity professionals. "Our data shows that AI is appearing most often alongside some of the strongest practitioners, not instead of them," Pylarinos stated. This distinction carries profound implications for security organisations and training institutions across Southeast Asia and globally, suggesting that investment in AI tools without corresponding investment in human expertise may prove counterproductive.
The transformation occurring within cybersecurity extends beyond competitive environments into broader industry practice. Pylarinos further noted that "as agents become more capable, human judgement, validation and hands-on technical skill become more important, not less." This observation contradicts fears prevalent in some quarters that automation and artificial intelligence will progressively eliminate the need for human decision-making in security operations. Rather, the evidence suggests an interdependence developing between machine learning systems and human expertise, where the value of skilled personnel increases as responsibility for validating and directing automated systems becomes more complex.
The emerging landscape also reflects a fundamental shift in cybersecurity's threat environment. Recent high-profile incidents, including Hugging Face's disclosure in July 2026 and the Open Web Application Security Project's (OWASP) compilation of generative AI exploits in the first quarter of 2026, demonstrate that artificial intelligence represents both opportunity and vulnerability. As organisations deploy AI defensively to identify and remediate threats faster, adversaries simultaneously weaponise artificial intelligence to develop novel attack vectors. This bidirectional transformation means cybersecurity leaders must contend with a landscape where intelligence technologies reshape both the attack surface and the defensive response simultaneously.
For security leaders navigating this transition, the strategic imperative centres on capability development rather than simple tool adoption. The challenge extends beyond purchasing or implementing artificial intelligence solutions; organisations must ensure their teams possess the foundational knowledge, technical acumen and analytical discipline to direct AI systems effectively, interrogate their conclusions and validate their outputs. This requirement places considerable emphasis on cybersecurity training and workforce development programmes, particularly across developing economies in Southeast Asia where talent shortages already constrain security operations.
HTB's research builds upon earlier work exploring controlled scenarios in which cybersecurity practitioners deliberately worked with artificial intelligence. The latest data represents a more organic picture of how security professionals choose to deploy these technologies when competing without external constraints or mandates. The progression from controlled experimental settings to observed real-world adoption patterns provides valuable insight into the genuine trajectory of artificial intelligence integration within professional cybersecurity operations, rather than theoretical possibilities or vendor marketing claims.
These findings carry particular resonance for Malaysian and Southeast Asian organisations grappling with cybersecurity capacity constraints. The region faces persistent challenges attracting and retaining cybersecurity talent, making the prospect of augmented capabilities through artificial intelligence particularly appealing. However, the HTB data suggests that successful implementation requires balancing technological investment with human expertise development. Organisations cannot simply deploy AI tools and expect performance improvements; they must cultivate teams capable of strategically directing these capabilities and maintaining oversight of automated decision-making processes.
The evidence further suggests that cybersecurity education and professional development must evolve to incorporate artificial intelligence literacy alongside traditional technical skills. Security professionals require not only technical expertise in their specialisations but also understanding of how artificial intelligence systems operate, their limitations, potential failure modes and ethical implications. This expanded skill set represents a significant departure from traditional cybersecurity career paths and creates both challenges and opportunities for training institutions and employers throughout the region.
Looking forward, the integration of artificial intelligence into cybersecurity operations appears irreversible and accelerating. The concentration of AI adoption among top-performing teams indicates that organisations not incorporating these capabilities risk falling behind competitors and threat actors alike. However, the research simultaneously underscores that technological superiority alone proves insufficient without the human expertise to direct, validate and contextualise machine-generated outputs. For Malaysian and regional security organisations, this reality necessitates parallel investments in both advanced tools and the development of skilled practitioners capable of extracting maximum value from artificial intelligence while maintaining the human judgement essential to modern cybersecurity operations.
