Revolutionizing AI and Data Security: A Unified Approach for Organizations
In a groundbreaking announcement, Proofpoint, Inc., a leader in cybersecurity, has introduced a revolutionary system designed to empower organizations to embrace artificial intelligence (AI) while maintaining control over their data and identifying emerging risks associated with both employees and AI agents. The newly unveiled Proofpoint Agentic Data and AI Security system serves as an integrated framework for managing AI and data as interconnected risks, ensuring that organizations can operate efficiently without compromised security.
This innovative system aims to simplify the complexities surrounding AI integration within business environments. It allows organizations to deploy AI agents that can access only the data necessary for their tasks, governed by specific business intentions. This aligns with existing corporate policies, translating them into real-time controls. By continuously monitoring AI behavior and data interactions, the system unveils potential risks that could arise from both human and AI activities.
One of the significant concerns currently facing organizations is that AI agents can rapidly discover, manipulate, and execute actions concerning sensitive data on an unprecedented scale. Traditional approaches often mean that AI tools may understand intent but overlook data access, whereas data security measures recognize sensitive data without insight into the agents’ intentions. This disparity can create blind spots, leaving organizations vulnerable to various risks. As the complexity and frequency of these interactions grow, the need for an agile system capable of responding to emerging threats becomes increasingly critical.
Moreover, the AI governance landscape has expanded beyond mere data loss mitigation. AI agents are now known to interact with enterprise systems, make autonomous decisions, and execute transactions, thereby introducing various operational, compliance, and safety risks. According to Proofpoint’s 2026 AI and Human Risk Landscape report, a staggering 87% of organizations have scaled AI assistants beyond initial pilot phases; however, over half remain uncertain about their existing controls’ ability to detect compromised systems. Given this reality, security teams must adapt to the urgency of defining and enforcing acceptable AI behaviors in real-time.
Mayank Choudhary, executive vice president and general manager of the Data Security and Governance Group at Proofpoint, emphasized the necessity of marrying data security with AI management. He stated, "You cannot secure AI without securing the data it acts on, and you cannot secure data without understanding how AI is using it." He argued that separating intent from access would result in valuable contextual information being disregarded. Recognizing the interplay between AI’s real-time operations and data governance equips organizations with insights to act swiftly in addressing risks.
The Agentic Data and AI Security system employs a novel framework that operates at the agent level. This system utilizes the same identity, data context, and access controls that have successfully protected the data of over 14,000 enterprises. At the heart of this system is the Proofpoint Knowledge Graph, which interrelates AI activities with data sensitivity and context, creating a comprehensive understanding of risk through Nexus models.
To enhance security, the system incorporates three new autonomous agents that collaborate based on a unified understanding of risk:
-
Zero-Touch Detection: This detection agent identifies intent and access as a cohesive signal, filtering out irrelevant anomalies and highlighting only the most critical actions.
-
Instant Investigation: The investigation agent automates the reconstruction of activities across data, identity, and behavior, radically reducing investigation time from days to mere minutes.
- Protection Optimization: The remediation agent translates investigative findings into actionable insights that encompass access remediation and Data Loss Prevention (DLP) policy optimization while including human governance for oversight.
Organizations today face the challenge of ensuring that established operational rules for AI utilization are diligently enforced while also identifying risks that may not have been previously recognized. Proofpoint’s Semantic Business Policies enable organizations to articulate existing governance policies in straightforward language, effectively enforcing rules that align with contemporary AI capabilities.
Such adaptations are paramount for keeping pace with the evolving AI landscape. Proofpoint’s Agentic Insights employ autonomous reasoning agents to analyze patterns of AI interactions and behaviors, thereby identifying emerging risks. Once validated, these insights can form the basis for new enforceable policies, ensuring that organizations remain ahead of potential threats.
In summary, the introduction of the Proofpoint Agentic Data and AI Security system signifies a paradigm shift in how organizations can engage AI within secure frameworks. By integrating detection, investigation, and remediation capabilities in one comprehensive system, organizations are positioned to manage the complexity of AI interactions while ensuring robust protections are in place. As Ryan Kalember, chief strategy officer at Proofpoint, aptly pointed out, "The challenge now is making those rules enforceable as AI takes on more consequential work."
For businesses looking towards the future, Proofpoint’s innovative solutions offer a promising pathway to secure AI operations and data control. The capabilities offered by the Proofpoint platform are expected to be available by year-end 2026, marking an exciting development in the field of AI and cybersecurity. As the landscape continues to evolve, organizations are reminded of the crucial need to stay informed and adaptable, ensuring that technology complements security, rather than undermines it.