Skip to content
2.X
2.76.0

Fides 2.76.0 release notes

  • 2.76.0: December 8, 2025
The Enterprise tag indicates that features are only available for Enterprise customers. To review pricing and upgrade your plan, please visit our site (opens in a new tab) or contact us to learn more.

Helios

Detect and classify global data risks with comprehensive visualization for data mapping and reporting for modern Enterprise.

  • System Management: The data stewards field in the system information form has been upgraded to a searchable multiselect component, making it significantly easier to assign and manage data stewards for your systems. The new interface provides a more intuitive search experience and streamlines the process of adding or removing multiple data stewards at once.
  • Detection and Discovery Enhancements (Beta): Significant improvements to the Action Center enhance visibility and usability for data classification workflows:
    • Confidence Level Visualization and Filtering: The Action Center now provides comprehensive visual indicators and filtering options for classification confidence levels. Users can quickly identify high-confidence classifications and prioritize their review efforts accordingly, helping teams focus on the classifications that need the most attention while expediting approval of high-confidence results.
    • Action Center Summary Cards: Added expandable confidence level summary cards to the Action Center home page that provide an at-a-glance view of your classification status across different confidence levels. This makes it easier to understand the state of your data discovery efforts and prioritize your review workflow.
    • Tree View Actions: Introduced direct "Remove" and "Classify" actions within the tree view, allowing users to take action on resources without leaving the tree navigation. This streamlines the classification workflow and reduces the number of clicks needed to manage discovered data.
    • Improved Status Indicators: Updated the tree status indicator component to provide clearer visual feedback about the state of resources in the discovery process.
  • Performance and Reliability Improvements:
    • Optimized Memory Usage: Significantly reduced memory consumption during promotion tasks for tables with large field counts. This optimization implements batch processing and selective column fetching, allowing Fides to handle discovery of extremely large databases without running into memory limitations.
    • Enhanced Tree View Completeness: The staged resource tree now displays empty databases, schemas, and tables, ensuring you have a complete view of your datastore structure. This makes it easier to understand the full hierarchy and provides better control over resource removal decisions.
    • Better Error Handling: Improved field-level error handling in LLM classification, capturing and reporting specific field errors instead of failing entire classification operations. This makes troubleshooting easier and prevents individual field issues from blocking the classification of other fields.
    • Confidence Rating Updates: Updated classification confidence bucketing logic to use a 1-5 confidence rating scale mapped to high, medium, and low confidence buckets, providing more granular confidence assessments for automated classifications.
    • BigQuery Sharded Table Detection: Refined sharded table detection to only apply to BigQuery monitors by default, preventing false positives in other database systems and improving the accuracy of discovered resources.
    • Monitor Task Reliability: Fixed a race condition that could incorrectly mark monitor tasks as "interrupted" even when they completed successfully, improving the reliability of long-running discovery tasks.

Lethe

Enterprise grade privacy request management and processing with a single orchestration layer for DSRs across vendors and systems

  • Privacy Center Security: Added SECURITY_HEADERS_MODE configuration to Privacy Center, allowing you to automatically set security best practice headers in responses. This improves the security posture of your Privacy Center deployment with minimal configuration.
  • SaaS Configurations: Added new API endpoints to retrieve SaaS integration configurations and dataset YAML files. This provides a foundation for programmatic template management and makes it easier to review and audit your SaaS connector configurations.
  • Manual Tasks Enhancements:
    • Customizable Email Templates: The External User Welcome email for manual tasks is now fully editable through the Admin UI messaging templates section. This gives you complete control over the communication sent to external users when manual tasks are assigned.
    • Conditional Task Dependencies: Manual tasks can now accept conditions on privacy request data including location and custom fields, enabling more sophisticated workflow automation based on request attributes.
  • DSR Execution Improvements:
    • Improved Duplicate Request Logging: Duplicate privacy requests now log with a "skipped" status instead of "error" status, providing more accurate reporting and preventing confusion in the UI.
    • Performance Optimization: Added a database index on providedidentity.privacy_request_id to eliminate full table scans when fetching privacy request identities, significantly reducing database load and improving DSR execution performance.

Janus

High performance consent recording and orchestration for data sharing, built for enterprise data engineering and AI pipelines.

  • v3 Consent API Development: Continued development of the next-generation consent APIs with several enhancements:
    • Historical Preferences Support: Fides can now accept historical privacy preferences via the received_at parameter on the /api/v1/privacy-preferences/pre-verified endpoint. This is particularly useful when migrating consent data from other systems, allowing you to preserve the original collection timestamps.
    • Enhanced Preference Metadata: Added source_preference metadata to the consent v3 preference's meta.fides object, providing better traceability of where preferences originated.
    • Improved Schema Organization: Updated consent v3 schemas so that collected_at is now a top-level field for easier access, and experience_config_history_id is nested under meta.fides for better logical organization.
    • User-Friendly Date Formats: Changed v3 privacy preferences pagination to use ISO 8601 date string format rather than raw milliseconds in filters, making the API more intuitive to use.
    • Better Validation: The current privacy preferences API endpoint now returns a proper 400 error when no identity is provided, giving clearer feedback about missing required parameters.
    • Performance Enhancement: Added a database index on privacy_preferences.created_at to improve query performance when filtering preferences by creation date.
    • Bug Fix: Fixed incorrect sync_session usage in v3 Consent APIs that could cause database session issues.

Fides core

  • OAuth Compliance: Fixed an issue where the /api/v1/oauth/token endpoint response was missing the required token_type field. The response now properly includes token_type: "Bearer", ensuring full compliance with OAuth 2.0 specifications.
  • Taxonomy: Updated custom fields to support custom taxonomies as locations or values, providing greater flexibility in how you organize and categorize your privacy-related data.

Integrations

  • Pagination Fix: Fixed an issue where the pagination controls on the Integrations page were displaying incorrectly, with buttons disabled or controls disappearing.
  • Iterable Integration Improvement: Fixed the test request for the Iterable integration, ensuring that connection tests work correctly and provide accurate feedback about integration health.

Database Migrations

This release includes two database migrations:

  • Index on providedidentity.privacy_request_id for improved DSR execution performance
  • Index on privacy_preferences.created_at for improved consent query performance

Both migrations use deferred index creation for large tables to minimize deployment impact.