AI Agents Enhance Data Loss Prevention Systems for Organizations
In a significant development in the realm of data loss prevention (DLP), a startup called Mind, spearheaded by Eran Barak, the former CEO of Hexadite, has successfully raised $72 million in a Series B funding round led by Crosspoint Capital Partners. This fresh infusion of capital aims to revolutionize DLP for enterprises increasingly reliant on artificial intelligence (AI) by transforming traditional, static systems into dynamic, interactive technologies.
Founded in 2023 and already boasting 85 employees, Mind has amassed a total of $112 million in funding, having previously raised $30 million in a Series A round in June 2025, also led by Paladin Capital Group and Crosspoint Capital Partners. Under Barak’s leadership since its inception, the company has evolved from its roots in machine learning to address the critical challenge of data protection in environments rich with AI applications.
Redefining Data Loss Prevention
The primary aim of Mind’s initiative is to shift DLP from a conventional methodology anchored in rigid classifications, policies, and blocking mechanisms to a more fluid, intelligent approach. Barak articulated that traditional endpoint controls would continue to manage the technical aspects of data protection, while AI agents would sift through substantial volumes of data lineage and activity, discerning what information truly requires scrutiny.
In Barak’s vision, Mind not only seeks to create a singular platform that integrates various forms of DLP—encompassing endpoint, SaaS, generative AI, and on-premises data—but also to unify these elements under one cohesive governance model. This complexity necessitates ample resources, yet, according to Barak, it is an essential evolution for the DLP field.
Streamlining Processes with AI
One of the notable challenges with traditional DLP systems is their tendency to consume significant resources at endpoints, often due to incessant scanning processes. Mind’s innovative technology aims to streamline these operations by determining when data needs to be scanned—utilizing previously established classifications to maximize efficiency. For instance, if Mind identifies classified information within services like Google Drive or OneDrive, it retains these labels when the information is transferred to an endpoint.
Barak pointed out the inherent friction and overhead that conventional DLP systems impose on users. Addressing this issue, Mind has prioritized creating a lightweight agent that efficiently covers a range of operating systems, thus simplifying the user experience.
Moreover, the startup is taking a sophisticated approach to data lineage. While existing systems can show activity such as file downloads, modifications, or personal identifiable information (PII) masking, Barak highlighted that merely presenting these details can overwhelm DLP analysts. The plan is for AI agents to autonomously analyze lineage data, presenting concise summaries of significant user actions, effectively allowing the technology to gauge the importance of user behavior.
Capturing Evidence Effectively
To improve the handling of sensitive data movements, Mind aspires to go beyond merely alerting users about data policy violations. The company envisions a system that captures substantial evidence of user actions, such as documenting instances where PII may have been inappropriately shared—say, sent to a personal email account. Barak underscored the difficulties posed by storage and compression challenges tied to parsing large volumes of information while retaining enough detail for investigatory purposes.
The company has made strides in developing a balance between the amount of data collected and the user experience, establishing a framework that narrates the user’s actions without overwhelming them. With the integration of AI agents, this process could become significantly more manageable.
Customized Endpoint Configurations
An innovative feature of Mind’s approach is its flexibility in configuring endpoints based on user roles and responsibilities. This adaptability allows employees to have clearer understandings of what sensitive information they are permitted to share, emphasizing guidance rather than rote blocking or allowing actions. In this model, AI agents interact with users, effectively educating them about policy violations and facilitating smoother user experiences.
Barak noted that customers can implement Mind’s technology swiftly—often within hours—and begin enforcing data protection measures using pre-defined policies, thereby avoiding cumbersome configurations typically associated with other DLP systems. The integrated nature of Mind’s offerings—spanning discovery, classification, policy, and prevention—sets it apart from competitors who often add functionalities piecemeal.
Future Developments and Challenges
Mind’s current focus lies on enhancing DLP for Windows and macOS, with future plans to tackle Linux and mobile platforms. The latter has emerged as a critical area of interest, given the growing reliance of sales and other field personnel on mobile devices, despite the inherent technical constraints these platforms pose.
As Mind embarks on its ambitious growth trajectory, having recently crossed eight-digit revenue figures, Barak remains steadfast in aiming for $100 million in annual recurring revenue within the next two years—a challenging yet achievable objective, given the pressing demand for advanced data protection solutions in today’s AI-driven landscape.
In conclusion, Mind’s innovative approach signifies a promising shift in the DLP landscape, combining the powers of AI with user-friendly functionalities to safeguard sensitive data while minimizing operational friction. The successful securing of $72 million in funding marks a pivotal point for the startup, reinforcing the role of technology in tackling the complex data security challenges enterprises face.