Aline’s new agentic intake turns request workflows into active AI-powered processes. Instead of just collecting information and waiting for teams to manually triage it, Aline AI reads the intake, reviews attached vendor materials, decides what work is needed, completes low-risk steps, and routes higher-risk items to the right legal, security, privacy, compliance, or procurement owner.

## Fixes

App menus now switch more smoothly when moving between menu items, reducing extra clicks.

Document/editor scrolling is now easier to use, with scrollbars that avoid overlapping resize handles and toolbar controls.

Repository uploads now show a more compact progress panel with clearer upload status, retry/delete/restore actions, and quick access to completed documents.

Fixed AI reminder generation failures caused by an invalid structured response schema.

Fixed an editor crash that could happen when inserting a new line in complex documents.

Fixed keyboard navigation in Extraction Question setup so Tab, Shift+Tab, and Enter move focus between fields instead of closing the menu.

Fixed Document Summary behavior for self-referential related documents by showing the actual document title and a “Document is already in view” notification.

Fixed a signing workflow issue where documents could be sent for signature even when fields referenced recipients who were no longer signers.

Fixed upload failures where local files could show “File not found,” especially after switching tabs or leaving uploads queued mid-session.

Improved Filestack upload handling so upload credentials stay fresh during long sessions.

This release includes several improvements to AI-powered reports that make answers easier to review, protect, and configure.

AI-generated report answers can now include a short explanation of why the answer was selected. This helps reviewers understand the source and logic behind an AI answer without manually re-reading the document from scratch.

You can now lock AI report answers so they are not overwritten by future AI runs. Individual cells, entire rows, or full columns can be locked depending on what you want to preserve. Locked answers show a clear lock indicator, and future AI runs skip those protected answers.

Report select questions now support adding options more easily from CSV-style input. This makes it faster to set up questions with many choices, especially when options already exist in a spreadsheet or list.

These releases also include follow-up fixes for report selection behavior and tab/modal stability, improving the overall experience when working with larger reports and bulk actions.

You can now : ask Aline to refine an existing AI redline, stack a second pass on the first, and see each change clearly in chat and in the document—without wiping prior suggestions or accidentally shifting neighboring contract language.

Select an AI suggestion in the document and use to open chat with that change in context. Tell Aline how to adjust the language (tighter indemnity, align with your playbook, respond to counterparty markup). The model works the prior redline, not as a blank-slate rewrite.

When you (or the AI) propose another edit where an AI redline already exists, Aline layers the new change correctly—e.g. first pass changes , a follow-up changes —with markers and text staying where they belong.

In the assistant, redlines stream in with views, including prior deletions, so you can sanity-check each pass before accept/reject. The suggestions toolbar also as redlines stream in, so you’re not waiting on the UI to catch up with the model.

We’re releasing a of our new AI Chat to select customers for testing. This upgrade is designed to make AI-assisted legal work more reliable, more transparent, and better suited to longer, more complex matters.

The new chat experience can now plan work before executing it, break larger requests into focused steps, and run multiple specialist agents in parallel. That means it is better equipped for longer-horizon work such as drafting document sets, comparing provisions across deal files, preparing summaries from multiple sources, and turning source materials into polished legal documents.

AI Chat now creates a structured plan for larger tasks, which helps it stay organized and complete longer workflows more consistently.

Multiple AI agents can now work at the same time on related tasks, such as drafting several definitive agreements from a term sheet or reviewing multiple documents in parallel.

Drafts now support richer formatting through HTML, including proper lists, page breaks, font styling, and cleaner document structure.

AI Chat can search across available matter or repository materials to help find relevant documents, clauses, and context before answering or drafting.

For complex tasks, the chat can show the steps being performed so users have a clearer sense of what the AI is doing.

You can ask AI Chat to help with tasks like:

This release is being made available in so selected customers can test the new experience on real workflows and give feedback. We expect to move this new AI Chat experience to general availability soon after testing.

Here is an overview video about improved legal drafting, parallel execution, and planning:

Here is an overview video about repository searching: