Malik Haidar is a veteran cybersecurity strategist who has spent decades navigating the complex digital defenses of the world’s largest multinational corporations. His career is defined by a rare ability to blend deep technical intelligence with high-level business logic, ensuring that security measures empower rather than hinder enterprise growth. Today, he shares his perspective on the rapid evolution of data governance and the critical need to secure the burgeoning ecosystem of autonomous AI agents.
Cyera recently reached a $12 billion valuation following a $400 million investment from Goldman Sachs Alternatives. How will this capital infusion specifically accelerate your platform’s expansion, and what key technological milestones do you aim to achieve with these new resources?
This $400 million investment, which brings our total funding to over $2.7 billion, serves as a powerful catalyst for our next phase of engineering breakthroughs. We are aggressively scaling our research and development to ensure our 1,500 employees can stay ahead of the increasingly sophisticated threats targeting global enterprises. The primary technological milestone is to deepen our real-time visibility capabilities, ensuring that as companies migrate to hybrid clouds, our platform remains the definitive source of truth for their data. By leveraging this capital, we can accelerate the rollout of automated remediation tools that act within seconds of a vulnerability being discovered.
Your platform utilizes an agentless, API-driven approach to provide real-time visibility across hybrid clouds, SaaS, and on-premises environments. Can you walk us through the technical process of leveraging machine learning to classify structured and unstructured data while maintaining business context and compliance?
Our agentless architecture is designed to eliminate the friction typically associated with legacy security deployments by connecting directly via APIs. We employ advanced machine learning models that scan everything from structured SQL databases to unstructured PDF documents to identify sensitive patterns. This process goes beyond simple keyword matching; it evaluates the metadata and surrounding context to determine if a file is a routine report or a high-stakes legal contract. By continuously mapping this data to specific compliance mandates like GDPR, we provide a live inventory that reflects the actual risk posture of the business at any given moment.
The integration of Data Security Posture Management (DSPM) with identity access governance allows for the discovery of over-permissioned access and misconfigurations. What are the most common high-risk sharing patterns you encounter in Fortune 500 environments, and how does correlating data sensitivity with user identity mitigate these threats?
In large-scale environments, we frequently observe “shadow data” where sensitive information is moved into unmanaged cloud buckets that lack proper oversight. A particularly dangerous pattern involves over-permissioned identities, where a single user or automated service has broad access to sensitive files that are entirely outside their job scope. By correlating the sensitivity of the data with the actual behavior and permissions of the user, we can instantly flag these misconfigurations. This unified approach allows security teams to enforce the principle of least privilege, effectively shrinking the attack surface before a malicious actor can exploit it.
As AI agents multiply within enterprise workflows, the gap between trusted actions and actual data access creates significant risk. How does your platform provide guardrails for LLM inputs and training datasets, and what specific steps are taken to prevent data exfiltration during automated AI operations?
The proliferation of AI agents has created a new frontier for data leakage that traditional security tools are simply not equipped to handle. Our platform implements rigorous guardrails by inspecting every piece of data intended for LLM training or input to ensure no intellectual property is exposed. We monitor these automated workflows in real-time, looking for anomalous movements that might suggest an AI agent is being used to exfiltrate data to unauthorized external endpoints. It is about creating a secure “sandbox” for innovation, where the speed of AI does not come at the expense of corporate confidentiality.
The $1 billion acquisition of Oasis Security signals a major move into agentic access management. How does this acquisition complement your existing data loss prevention strategies, and what practical advantages does it offer to organizations struggling to track the permissions of autonomous AI agents?
Acquiring Oasis Security for $1 billion allows us to solve the most pressing challenge of the modern ermanaging non-human identities. While traditional data loss prevention focuses on what people do, this acquisition gives us the tools to govern what autonomous AI agents are doing behind the scenes. It provides a structured framework to manage the permissions of these “digital workers,” ensuring they only interact with the data they are strictly authorized to use. For organizations struggling with the sheer scale of AI automation, this offers a clear, manageable view of every agentic action within their network.
With over 1,500 employees and a client base that includes 20% of the Fortune 500, the scale of operations is significant. Could you share a scenario where a large-scale enterprise successfully unified its data security across disparate environments using your platform, including the specific metrics used to measure their security improvement?
One of our clients in the global manufacturing sector recently unified their entire security stack across three different cloud providers and numerous on-premises data centers using our platform. Within the first few weeks, they identified and secured over 50,000 sensitive files that were previously “invisible” to their legacy systems. They measured their success through a 75% reduction in time-to-remediation for critical vulnerabilities and a significant decrease in manual audit hours. This transformation proved that even the most fragmented environments can achieve a cohesive and measurable security posture with the right visibility tools.
What is your forecast for the future of data security in an era dominated by autonomous AI agents and hybrid cloud complexity?
From 2026 to 2028, the industry will pivot toward a “data-first” identity model where the security policy travels with the data itself, regardless of whether a human or an AI agent is accessing it. We will see the total disappearance of the traditional network perimeter, replaced by intelligent, self-healing data environments that can detect and block unauthorized access in milliseconds. The most successful enterprises will be those that embrace automated governance, allowing them to scale their AI operations without the constant fear of a catastrophic data breach. Ultimately, security will become an invisible but unbreakable fabric that supports every digital interaction within the global economy.
