New AI-aware data controls
Commvault added real‑time controls for structured and vector databases to curb AI-driven data leakage and reduce inadvertent PII exposure as 90% of firms expose AI-visible data. The move comes as the NSA issues fresh guidance on AI supply‑chain risks, highlighting the need for provenance, continuous auditing and third‑party risk checks across ML stacks. (stocktitan.net) (dig.watch)
Commvault published the update for Commvault Cloud on March 18, 2026 as a PR distributed via PR Newswire and the company newsroom. (prnewswire.com) The Satori deal that enables the new controls closed on August 28, 2025, with Commvault reporting a total cash consideration of $28.3 million for Satori Cyber Ltd. (marketscreener.com) Commvault’s release specifies the expansion embeds Data Security Posture Management (DSPM) into structured data environments and adds real‑time Data Access Governance that targets structured databases and vector stores while surfacing LLM monitoring and prompt protection capabilities. (commvault.com) Product collateral and the Commvault AI pages frame these controls as integrated into cyber‑recovery workflows with immutable backups, granular recovery and policy‑based governance across clouds to support trusted reuse of training data. (commvault.com) Independent telemetry cited by vendors shows the scale of the exposure Commvault is addressing: Varonis’ 2025 State of Data Security report analyzed 1,000 environments and reported 99% of organizations had sensitive data exposed to AI tools and that 90% of sensitive cloud data was accessible to AI. (info.varonis.com) The NSA and allied agencies released joint AI/ML supply‑chain guidance in early March 2026 that defines six supply‑chain components (training data, models, software, infrastructure, hardware, third‑party services) and recommends provenance tracking, verified model registries, AI/software bills of materials, integrity checks and improved third‑party visibility. ( )
