AI Security: "Coin Toss" Effectiveness Amid Rapid Adoption

AI tools are widely and quickly deployed, but security is struggling to keep up. This gap is no longer theoretical; it is already creating real-world risk. A new report finds that 87% of organizations have AI assistants beyond pilot stages, meaning AI is embedded in daily work. However, only 63% have AI-specific security controls. Even then, confidence is low, with 52% not fully trusting their controls to detect compromise. Among organizations with AI security controls, 50% reported an AI-related incident. This means it's roughly a coin toss whether controls prevent or miss incidents. Only 33% feel fully prepared to investigate AI incidents, and 95% struggle with managing multiple security tools. This complexity makes it harder to see what's happening in AI-enabled workflows. This matters because the rapid adoption of AI is creating immediate and measurable security risks that organizations need to address now.