AI Now Both Weapon and Target in Cyberattacks, CrowdStrike Report Finds
New data shows AI-generated signals now outpace human activity as attackers target the models themselves.
CrowdStrike has released its 2026 Threat Hunting Report, warning that artificial intelligence has evolved into both a primary tool for cyberattacks and a high-value target for adversaries. The findings signal a critical shift in the cybersecurity landscape where AI infrastructure now requires the same level of protection as traditional IT systems.
According to the report, the scale of AI-driven activity has surged, with AI-generated signals now exceeding human-generated activity. CrowdStrike identifies a ratio of 2.5 AI-generated signals for every single human-generated signal. To manage this volume, the company's threat hunting systems analyze more than 14 million detection leads every day. These massive datasets are filtered down to approximately 36,000 daily incidents that are reported to customers as likely malicious activity.
The AI Integration Trend
This shift occurs as organizations globally integrate AI to automate their defensive security operations. However, this adoption has created a parallel evolution in offensive tactics. Adversaries are no longer just using AI to scale the volume and speed of their attacks; they are now specifically targeting AI models. By compromising the integrity of these models or the data feeding them, attackers can potentially bypass security controls or manipulate system outputs.
Implications for the Industry
The transition of AI into the role of both "weapon and target" marks a new phase of cyber warfare. Because AI is being used to automate the discovery of vulnerabilities and the execution of exploits, the window for human defenders to respond has shrunk. This environment underscores an urgent need for AI-native security solutions—tools designed specifically to detect and neutralize AI-driven threats that traditional signature-based or heuristic defenses may miss.
Future Outlook
As AI continues to permeate corporate infrastructure, the security of the AI pipeline—from training data to model deployment—will become a primary focal point for risk management. Industry observers will be watching to see if the gap between AI-generated offensive signals and human-led defense continues to widen, and whether new standards for "AI hardening" emerge to protect models from being turned against their owners.