AI-Driven Threats Targeting U.S. Organizations

The report analyzes how adversaries are embedding AI into cyber operations against U.S. organizations to accelerate social engineering, identity compromise, vulnerability discovery, and malware variant creation, citing 2025 sources (OpenAI, Microsoft Digital Defense Report, Cloud of War) with data points such as 28% of breaches tied to phishing/social engineering, 18% to unpatched web assets, a 32% rise in identity-based attacks, and a 2,200% increase in AI-assisted attacks. It highlights heightened risk to critical infrastructure and cloud environments due to automation and agentic capabilities, and urges defenders to operationalize AI for behavioral detection, exposure management, and automated response, concluding with a recommendation of SOCRadar’s Agentic Threat Intelligence.