Draft 3.8: draft in English, deliver in 38 languages
Draft 3.8 produces a delivery-language version alongside the working draft, with locked defined terms, a parity check on every clause, and a confidence tier per language.
Review 4.1: governing-law and jurisdiction mismatch detection
Review 4.1 reads governing law, forum, arbitration seat and the schedules together and flags where they disagree. Also: Counsel in the background pass, and quieter Note flags.
Recall 2.6: citations that survive translation
Recall 2.6 anchors every citation to the source-language clause, so a pinpoint reference stays valid whichever of the 38 languages you read the answer in.
Trial & Arbitration Simulation is generally available
After a year in limited release, Simulation is open to all accounts: an AI tribunal, opposing counsel and witnesses, scorecards you can argue with, and 22 venues modelled.
Review 4.0: learning from rejections inside the matter
Review 4.0 remembers which redlines you declined on this matter and stops proposing them. Nothing leaves the matter, and nothing reaches the playbook unless its owner puts it there.
Draft 3.6: the deal-point form, and why boilerplate should write itself
Draft 3.6 introduces the deal-point form: answer the twenty or so questions that actually vary between deals, and Draft composes the rest from the playbook and voice profile. What it asks, what it assumes, where to be careful.
Recall 2.5: the trace view, or how to show your working
Recall 2.5 adds the trace view: for every answer, the exact passages it rests on, the documents they came from, the permission path that allowed them, and the steps between question and answer.
AI Interview is generally available: the first conversation, handled with care
AI Interview moves from limited preview to general availability. It conducts a first client-intake conversation, by voice or text, in 38 languages, and produces a structured brief with limitation dates and a conflict check.
Review 3.8: the Word add-in rebuilt on the native comments API
Review 3.8 replaces our custom annotation layer in Word with the native comments and tracked-changes APIs. Flags are now real Word comments; redlines are real tracked changes. What that fixes, and what it costs.
Draft 3.4: voice modelling on your last 200 closed deals
Draft 3.4 learns a firm’s drafting voice from its own closed deals: defined-term conventions, sentence rhythm, preferred constructions. What it learns, what it ignores, and how to check its work.