What the Largest Controlled Study of AI Attackers Reveals About Cyber Deception

Cyber deception was built to fool human attackers. Honeypots, honeytokens, and decoys rely on assumptions about how adversaries recognize risk, prioritize targets, and respond to suspicious environments.

But autonomous, LLM-driven attackers don’t behave like humans.

So, does deception still work?

Horizon3.ai researchers tested 21 AI models across 10 providers, analyzing 10,962 attacker decisions and benchmarking their behavior against 47 human red-teamers.

The findings challenge decades of conventional thinking about cyber deception.

AI attackers took the bait more than twice as often as humans.

Even more surprising: advanced models frequently recognized a trap in their own reasoning and attacked it anyway.

The result is a fundamental shift in how security teams should think about deception.

Inside the Whitepaper

Learn:

• Why AI attackers fall for cyber deception at significantly higher rates than human attackers.
• How the “recognition-action gap” causes AI models to identify traps and attack them anyway.
• Which traditional deception assumptions break when applied to autonomous attackers.
• Why decoys may no longer reliably divert attackers from real assets.
• How honeytokens and canaries can become high-yield early-warning signals for AI-enabled attacks.
• Why deception strategies should shift from misdirection to detection.
• How security teams can adapt defensive programs for frontier models and self-hosted AI agents.

The study also compares AI and human behavior across file systems, .htaccess files, HTTP responses, and HTTP requests, finding AI attackers more likely to take planted bait across every tested artifact category.

Who Should Read This

This whitepaper is designed for:

• Chief Information Security Officers (CISOs)
• Security Architects
• Threat Detection and Response Leaders
• Security Operations and Engineering Teams
• Threat Intelligence and Active Defense Teams
• Security leaders preparing for AI-enabled and autonomous attackers

Whether you’re already using cyber deception or evaluating how your defensive strategy must evolve for AI-driven threats, this research provides practical guidance grounded in observed attacker behavior.

Download the Whitepaper

AI attackers are more capable at finding real vulnerabilities. They’re also markedly easier to catch in the act.

The question isn’t whether deception still works.

It’s whether your deception strategy is designed for the attacker that’s coming next.

Download Hacking the Hackers: Can You Still Deceive an AI Attacker? and learn why security teams must rethink deception for autonomous, AI-driven adversaries.