SEC536: Adversarial AI - Penetration Testing AI Systems


Experience SANS training through course previews.
Learn MoreLet us help.
Contact usBecome a member for instant access to our free resources.
Sign UpWe're here to help.
Contact UsKey findings:
The survey's overall pattern points to a gap between the theoretical value of cyber deception and its practical adoption, driven primarily by non-technical, perception-based barriers rather than genuine technical limitations. The paper's hands-on lab test directly challenges the "implementation complexity" barrier that respondents rated most significant, demonstrating that a basic deception capability can be stood up quickly by someone with limited prior experience using free, open-source tools. Respondents were cybersecurity and IT professionals from medium-to-large enterprises across multiple industry sectors, recruited via LinkedIn and direct outreach.
Research found the primary barriers are non-technical rather than technical: perceived implementation complexity (rated 3.95 out of 5), lack of awareness (3.90), and unclear return on investment (3.60) all outweighed actual technical hurdles like integration challenges.
In a hands-on test, the open-source tool OpenCanary was deployed and validated with a working honeypot in under 30 minutes, using just 12 commands and no debugging or errors.
Yes. Third-party research cited in the paper found organizations using cyber deception reduced attacker dwell time from 277 days to 181 days, a reduction of nearly three months.
CISOs and Security Operations leaders were cited most frequently as primary decision-makers, while CIOs and Compliance Leads were mentioned less often, suggesting these tools are often treated as discretionary operational enhancements rather than compliance mandates.
Regulatory incentives rated highest among enablers (4.3 out of 5), followed by clearer return on investment (4.2) and greater access to real-world case studies (4.1).
















