Features can now be canceled without deploying, and the release adds an Investment Analysis app, deployment autonomy settings, smarter voice and agent tools, and fewer stalled runs.

A weekly social draft queue, voice relay to coding agents, and automatic meeting minutes arrive alongside more reliable agent runs and more accurate billing.

Document editing is safer against data loss, sessions recover automatically after reboots or disconnects, credit alerts are clearer, and two longstanding bugs around task duplication and PR status are fixed.

Meta's Muse Code joins as a supported coding agent, auto-reload limits rise to $10,000, and chat handles unsupported models and flaky AI responses more gracefully.

Setup is now a guided checklist, a new Spend Guard keeps run costs under control, computers that go offline mid-run now fail over automatically, and video, image, and voice tools got a round of model and reliability fixes.

AIExpedite added Meta and Muse model support, faster computer onboarding, real-time iOS voice chat, and a video editor assistant that can edit storyboards directly, alongside workspace, deployment, and reliability fixes.

Coding agent sessions are more resilient, voice triage got smarter, several member-management bugs are fixed, and release notes can now be published to X.

This blog now has a Changelog: the release notes for AI Expedite, grouped by theme and linked to the work behind them.