Cisco Warns of Widening AI Security Gap as Real-World Attacks Surge and Enterprises Struggle to Keep Up

Cisco's State of AI Security 2026 report warns of a surging AI threat landscape where research-lab vulnerabilities are now fueling real-world attacks, exposing a critical readiness gap as 83% of enterprises plan agentic AI deployments but only 29% feel equipped to secure them.

• Cisco releases its State of AI Security 2026 report, highlighting a rapidly expanding AI threat landscape where vulnerabilities once confined to research labs are now materializing in real-world attacks, including AI-enabled malicious campaigns observed in the second half of 2025.
• The report reveals a dangerous gap in enterprise AI readiness, noting that while 83% of organizations planned to deploy agentic AI, only 29% felt prepared to do so securely, with risks growing around prompt injection attacks, AI supply chain vulnerabilities, and autonomous AI agents operating across critical workflows.
• Global AI policy is shifting away from safety-focused regulation toward innovation and investment, with the US, EU, and China each taking distinct approaches, while Cisco's research team continues releasing open-source security tools to address emerging threats like MCP agentic AI risks and open-weight model vulnerabilities.