2.1.0

This release contains updates and improvements to Epiq AI. The following notes highlight new enhancements to Epiq AI and provide a list of bugs addressed in this release.

​

Release Highlights

• Epiq AI for Privilege Review Enhancements. Privilege Review now includes significant usability and workflow improvements, including an enhanced user experience, smarter domain analysis and entity labeling powered by case-specific data, field-level synchronization between libraries and review objects, and the ability to run Privilege Review workflows on selected datasets for greater flexibility and control.
• Review Run Enhancements. Review Runs now provide greater transparency, efficiency, and flexibility with visible relevance scores, customizable export options, bulk prediction and tagging operations, intelligent job queue estimates, multi-modal document analysis for text and images, and automated output validation to improve review accuracy and reduce manual effort.
• Issue Review Workflow. Epiq AI now includes an Issue Review workflow template that enables teams to apply custom review protocols at scale. Define your own issue tags, coding categories, and document-level fields, then build and run workflows that consistently classify and capture review data across large document populations.
• PII/PHI Review Workflow. Epiq AI now includes a PII/PHI Review workflow template that helps teams identify and analyze personally identifiable information (PII) and protected health information (PHI) at scale. Using a standard privacy-focused taxonomy, the workflow automatically detects sensitive information, associates it with affected individuals, and supports consistent, defensible reporting for data breach and incident response reviews.

​

New Features in Release 2.1.0

This release contains the following new features.

​

Issue Review Workflow

Epiq AI’s Issue Review workflow applies a review protocol to a document population: you define the tags the documents should be coded against, and the fields that should be captured for each document, and the workflow applies them consistently at scale. Unlike the Privilege and PII/PHI workflows, which ship with a fixed set of components and a purpose-built taxonomy, Issue Review is built around whatever catalogue your matter calls for — issue coding, contract clauses, custom review criteria, or any tabular review where the categories are yours to define. This workflow is available as a template that starts with a single classification step, which you then build out. You add steps as the review requires, run individual steps independently to execute one pass in isolation, or deploy the full workflow to run every step in sequence. Because the tags and fields come from your protocol rather than from a product taxonomy, the same workflow shape supports a two-tag relevance pass and a fifty-issue coding exercise equally well.

​

PII/PHI Review Workflow

Epiq AI’s PII/PHI Review workflow streamlines privacy and data breach review by automating the detection of personally identifiable information (PII) and protected health information (PHI), and by identifying the individuals that information belongs to. Using AI-driven analysis of document text, it helps teams apply a consistent element taxonomy across large datasets and produce defensible, reportable counts of what was exposed and who was affected. The workflow is optimized for incident response, where scope and notification obligations have to be established quickly and evidenced later. This workflow is available as a template optimized for PII/PHI review, allowing you to get started quickly using a proven configuration that includes a standard PII and PHI element taxonomy. You can run individual components independently to execute a specific analysis step in isolation, or deploy the full workflow to run both components in sequence using the current configuration. This flexibility supports a range of review strategies, from a scoping pass over a sample set to a full end-to-end privacy review.

​

Enhancements in Release 2.1.0

This release contains the following enhancements and updates.

​

Review Run enhancements

Review Runs now provide greater visibility, flexibility, and efficiency for AI-powered document review. You can view, sort, and filter results by relevance score, customize exports to include scores and reasoning, and perform bulk actions to apply predictions and update tags across multiple documents. Review Runs also support multimodal analysis using vision-enabled AI models to evaluate documents containing both text and images. Additional improvements include automatic validation of review outputs to identify and populate missing tags, as well as dynamic completion time estimates for Review Runs. Key benefits:

• Prioritize validation efforts using relevance scores.
• Export only the review data needed for downstream analysis.
• Reduce manual effort with bulk tagging and prediction actions.
• Analyze text- and image-based content within a single review workflow.
• Improve output consistency through automated validation.
• Gain better visibility into review processing status and expected completion times.

​

Epiq AI for Privilege Review enhancements

​

Workflow Experience Improvements

The Privilege workflow has been redesigned to simplify navigation and provide a more consistent user experience. Configuration and Flow have been consolidated into a single Steps tab that presents workflow stages in a numbered sequence and centralizes workflow configuration, execution, and results. Additional interface updates across workflow components improve visibility of step status and results, streamline input and output navigation, and enhance overall workflow usability.

​

Multimodal Processing for Privilege Workflows

Privilege workflows now support multimodal processing, enabling vision models to analyze documents that contain both text and images. Users can configure processing by file type and selectively route supported files through a multimodal model while continuing to process other files using traditional text extraction. This enhancement improves analysis of image-based and visually rich documents that may contain privilege-related information.

​

Enhanced Entity Labeling and Data Enrichment

Entity Labeling now includes enhanced domain analysis capabilities that automatically assign and enrich domain information using case-specific data. New domain and domain alias information is available for identified entities, providing additional context and improving domain classification accuracy. This release also introduces field-level synchronization between libraries and entities, allowing users to selectively enrich entity records with specific library fields. Additional reporting enhancements include tag-based analysis within Privilege Classification Metrics.

​

Model and tagging enhancements

• Include list of documents that are not scored
• Support export of model scores with 2 decimal points
• Increased the maximum number of tags for text generation from 50 to 100.
• Added support for aggregated model predicted tags.

​

Metadata and overlay handling

• Improved metadata propagation across overlays, including:
  • doc_date
  • doc_name
  • Support for doc_family_id overlays
• Updated privileged metadata handling to retrieve the most recent overlay metadata.

​

Technical Support update

Updated the Technical Support form to improve issue reporting and support request tracking.

​

UI and interaction improvements

• Added auto-scroll to dropdown lists that use pagination, enabling users to load additional results as they scroll.
• Reduced delay when displaying full text on hover in search filters.
• Improved visibility of selected values to prevent truncation.

​

Bugs fixed in Release 2.1.0

The following bugs were resolved in this version.

Number	Description
AIDA-1917	Fixed an issue where privileged metadata did not reflect the latest overlay values by updating logic to retrieve the most recent overlay metadata.
AIDA-1919	Fixed inconsistent propagation of overlay metadata by adding support for doc_date, doc_name, and doc_family_id.
AIDA-1655	Fixed an issue where the upload metadata table displayed empty values due to header mismatch by aligning header field names with internal field IDs.
AIDA-1916	Fixed an issue where text generation results saved list values incorrectly when is_list was false by converting values to strings.
AIDA-1909	Fixed an issue where predicted_tags returned as null despite existing data by preventing null values from overwriting generated tags.
AIDA-1918	Fixed a dtSearch error when processing foreign-language name searches.
AIDA-1902	Fixed an issue where the upload name field was not editable for pre-processed documents and native overlays by restoring editability and enforcing a valid name requirement.
AIDA-1910	Fixed an issue preventing user account creation when an email address contained an apostrophe.
AIDA-1612	Fixed an issue where dropdowns with pagination could not load additional results by adding auto-scroll behavior.
AIDA-1840	Fixed an issue with delayed hover text and truncated values in search filters by reducing hover delay and improving visibility.
AIDA-1893	Fixed an issue where the translation feature was missing a scrollbar and experienced delays by restoring the scrollbar and improving responsiveness.
AIDA-1689	Fixed an issue where certain endpoints lacked validation and could expose files by adding missing sanity checks.