A recent Wiz survey reveals a significant gap between the rapid adoption of cloud-based AI services and the security measures in place to protect them, with 87% of organizations already using tools like OpenAI or Amazon Bedrock, but only 13% implementing AI-specific security controls, reports Virtualization Review.
The report, AI Security Readiness: Insights from 100 Cloud Architects, Engineers, and Security Leaders, highlights how security teams are often tasked with defending systems they don’t fully understand, with 31% citing a lack of AI security expertise as their top concern. Despite the prevalence of hybrid and multi-cloud deployments, many firms still depend on legacy tools like endpoint detection and response, which fall short against AI-driven threats such as poisoned training data or unauthorized API use. Wiz outlines a four-stage AI security maturity model and calls for continuous discovery, AI-specific policies, and integrated DevOps workflows. The report concludes that proactive and automated security strategies are now essential for protecting AI workloads.
