As organizations increasingly rely on telemetry and observability data to power operations and AI workflows, protecting sensitive information inside logs has become a growing challenge. A new partnership between two data infrastructure vendors aims to solve that problem without sacrificing usability.

Observability platform provider Coralogix and data privacy platform Skyflow announced a strategic collaboration designed to safeguard sensitive customer data embedded in logs while maintaining the full functionality needed for debugging, security analysis, and AI-driven operations.

The joint solution introduces a privacy-first observability model that replaces traditional redaction techniques with tokenization-based data protection, allowing organizations to keep telemetry data searchable and usable without exposing sensitive information.

The Hidden Risk Inside Observability Data

Logs and telemetry are foundational to modern software operations. Engineering teams depend on them for diagnosing application issues, investigating incidents, and monitoring system health.

However, these data streams frequently contain sensitive information—such as customer identifiers, account numbers, or personal details—hidden within both structured data fields and unstructured log messages.

Most observability platforms address the issue by masking or removing sensitive data entirely. While that protects privacy, it often reduces the usefulness of logs.

Redacted logs can create operational problems such as:

• Broken identifiers across events
• Reduced search and correlation accuracy
• Limited context for security investigations
• Reduced effectiveness of AI-driven monitoring tools

According to Anshu Sharma, the traditional approach forces organizations into a difficult compromise.

“Once sensitive data is stripped out, teams lose the ability to search effectively, investigate incidents, or let AI agents reason over what actually happened,” Sharma said.

Tokenization Instead of Redaction

The Coralogix–Skyflow integration takes a different approach by using privacy-preserving tokenization rather than removing sensitive data entirely.

Instead of masking values permanently, Skyflow replaces sensitive data elements with consistent tokens. These tokens maintain the structural integrity of the data, enabling engineers and analytics systems to continue searching, correlating, and analyzing events.

The original sensitive values are stored separately within Skyflow’s secure environment and can only be accessed through controlled, policy-based permissions.

This approach ensures that observability data remains operationally useful while sensitive information remains protected.

For Ariel Assaraf, the partnership addresses a growing challenge for organizations relying on observability data as a system of record.

“Customers shouldn’t have to choose between safeguarding sensitive data and maintaining operational efficiency,” Assaraf said.

Supporting AI-Driven Observability Workflows

The collaboration is also designed to support the rising use of AI in observability systems.

AI-powered monitoring tools increasingly analyze telemetry data to detect anomalies, predict incidents, and automate remediation processes. However, exposing raw customer data to these systems introduces compliance and privacy risks.

With the tokenization model, AI tools can still analyze logs and telemetry without accessing the underlying sensitive data.

Key capabilities of the joint solution include:

• Keeping sensitive customer data out of logs, dashboards, and analytics tools
• Maintaining search, filtering, and event correlation capabilities
• Allowing AI agents to analyze telemetry safely without direct data exposure
• Enabling policy-based rehydration of sensitive data for authorized workflows
• Reducing data sprawl across observability pipelines

The result is an observability environment that supports both AI-driven automation and strong data governance.

Addressing Data Sovereignty Requirements

Global organizations also face increasing regulatory requirements around data residency and sovereignty.

Coralogix already allows customers to deploy observability workloads in specific geographic regions, helping organizations meet regional data compliance standards.

By integrating Skyflow’s runtime data control capabilities, companies can extend that model further—ensuring sensitive customer data remains isolated, access-governed, and auditable while operational telemetry stays locally accessible.

This architecture helps organizations operating across multiple jurisdictions minimize cross-border data exposure while maintaining operational visibility.

Observability for a Privacy-First Era

The rise of AI-driven infrastructure monitoring, automated operations, and data-heavy applications has made observability more important than ever.

But it has also increased the risk of sensitive information appearing in operational data streams.

The partnership between Coralogix and Skyflow reflects a broader shift toward privacy-by-design observability systems, where security and operational usability are built into the data pipeline from the start.

By combining observability analytics with privacy-preserving data controls, the companies aim to provide organizations with a new model for managing telemetry data in an AI-powered world.

Get in touch with our MarTech Experts.