Commvault announced new upcoming AI capabilities that will enable organizations to embrace AI while maintaining control of their data, agents, and recovery.
These capabilities enable enterprises to securely activate AI, discover and manage AI agents, and build and control agent workflows, all from Commvault Cloud.
As organizations implement AI, many struggle to balance innovation with the potential for leakage of sensitive data and loss of control over AI agents. According to Deloitte, 60% of AI leaders cite risk and compliance concerns and legacy system integration as the biggest barriers to agentic AI adoption. Building on AI resiliency, Data Activate, AI Protect, and AI Studio address these challenges by helping teams understand the impact of agent-driven changes and rollback when necessary.
Sanjay Mirchandani, President and CEO of Commvault, said: “Every era of enterprise has created systems of record: ERP for business operations, CRM for customers, and now AI for enterprises.” “When the data supporting AI is compromised, the AI is compromised. If you can’t recover the data, you can’t trust the AI. Commvault Cloud is the system of record for AI resiliency.”
Data Activate allows organizations to classify and curate data from protected backup copies and prepare approved datasets in formats like Apache Iceberg and Parquet for use in large-scale language models and AI data platforms. The product continually publishes updated and vetted datasets, making it easy to keep your AI pipelines and trusted data in sync. For example, teams can identify and exclude personally identifiable information before activating a dataset for model development. Built on Commvault Cloud’s governance and Zero Trust architecture, Data Activate enables secure and controlled data activation while mitigating the potential exposure of sensitive data through AI.
AI Protect helps organizations identify vulnerabilities, understand the impact of agent-driven changes, recover affected applications, and perform full-stack recovery across AI-driven environments. As application stacks evolve and more agents are able to perform unrestricted unwanted changes across data, systems, and configurations, organizations need a unified approach to managing both complexity and threats. AI Protect discovers and inventories agents across your environment and maps their activity to your AI stack.
AI Studio enables organizations to create and use agents for their unique needs, including a repository of built-in agents for common resiliency use cases. Teams can now automatically and securely use Commvault’s Model Context Protocol server to build custom agents that integrate with other enterprise systems.
With these innovations, the team continues to advance the AI resiliency lifecycle, starting with securely activating data for AI purposes and ending with discovery, governance, protection, and recovery for AI agents across on-premises, SaaS, and hybrid cloud environments.
“When deploying AI agents throughout an organization, the primary concern will be maintaining visibility and control over what these agents have access to and how they interact with sensitive data,” said Marius Hoja, CCoE Computing Architecture and Engineering at Emerson Electric. “The ability to view, manage, govern, and orchestrate your agent ecosystem in real time from a single platform gives you even more confidence to scale your AI innovations without sacrificing safety or resiliency.”
“As enterprises accelerate AI adoption, they need solutions that reduce risk while delivering real business value,” said Sean Alexander, SVP Connected Ecosystems at Lumen. “We bring together secure, scalable connectivity to help our customers deploy with confidence and innovate securely in the age of agents.”
Pranay Ahlawat, chief technology and AI officer at Commvault, said, “In an agent environment, agents change the state of data, systems, and entire configurations in ways that are rapidly compounded and hard to track. When something goes wrong, teams need to restore not just the data, but the entire stack (applications, agent configurations, and dependencies) to a known good state. That’s what AI Protect provides.”
