Welcome to Ada’s release notes. Scroll down to see a list of recent releases, or subscribe to get notified about updates.
Subscribe via email
Receive a weekly email summary of releases every Friday at 11 a.m. Eastern, provided there has been at least one release that week.
Subscribe via RSS
Copy the following URL into your RSS reader to get notified about new releases:
Topic and Intent classifications in Data Export
The Conversations endpoint in the Data Export API (v2) now exposes a new classifications field. The field returns the Topic and Intent classifications for a conversation. Each entry includes a…
Topics and Intents
Each Topic groups conversations by subject and contains Intents that capture the specific reason behind a conversation. The Analytics > Topics & Intents page shows volume, AR rate, AR opportunity,…
AI-recommended Intents
When conversations don't fit the existing taxonomy, recommended Intents and Topics are surfaced for review. Review a suggestion's volume and trend, then approve it into the taxonomy or dismiss it.
Faster classification
Open conversations are classified about 10 minutes after the end user's last message and reclassified when new activity arrives, rather than only after a conversation ends. Conversations handed off…
Upgrading from Topics
A draft taxonomy is generated from historical conversation data. Review and edit it on the Topics & Intents page, then upgrade to replace legacy Topics. Upgrading is final; legacy Topic data remains…
What changed
reset continues to apply greeting and language and to clear chat history. Any metaFields or sensitiveMetaFields passed to reset are now ignored unless the new allowMetaFieldsInReset setting is set to…
Who it affects
Integrations that rely on passing meta fields through reset must add allowMetaFieldsInReset: true to their Web Chat settings to keep that behavior. Most integrations set meta fields through the…
Why it matters
Meta fields are sometimes used to drive entitlements such as discounts, vouchers, or account benefits. Ignoring them on reset by default prevents end users from altering those values from the…
Filter conversations by moderation category
The Convos view now supports a Moderation filter that surfaces conversations where the moderation filter triggered. Previously, identifying potentially unsafe conversations required parsing…
Redesigned Conversation View
The Conversation View has a refreshed design that makes conversations easier to find, scan, and review. Conversation cards now lead with the conversation topic and a short summary of what the end…
Genesys handoff integration
AI Agent conversations can now be escalated to a human agent in Genesys Cloud. The integration creates a new Open Messaging conversation routed through your Inbound Message flow and bidirectionally…
What's changing
Playbooks are now built around a structured step-based model that lets you define exactly how your Agent executes each phase of a conversation. The legacy freeform authoring format is now called…
More control over each step
Every Playbook is built from six discrete step types, each performing a single, well-defined operation: SEND: Send a message to the end user (Fixed message or Contextual AI-generated). SET: Assign a…
Built-in flow intelligence
Playbooks now monitor conversations during execution and intervene automatically: Intent change detection: Exits the Playbook and reroutes when the end user's intent shifts. Pause and resume for…
Action error handling
When an Action returns a 4xx or 5xx HTTP status code, save the status code to a variable and branch on it with IF/ELSE to build custom fallback behavior.
Multi-channel support
Playbooks are supported on Messaging, Email, and Voice channels. Learn more in the Playbooks documentation.
MCP: create coaching with propose_change
propose_change can now create new coaching, not just update existing coaching. AI assistants can analyze a conversation, identify the specific AI Agent turn that went wrong, and author coaching…
Control multilingual Knowledge ingestion
AI managers can now control whether their AI Agent ingests non-English Knowledge directly from the dashboard, using the new Multilingual Knowledge toggle in the header of the Knowledge page. This…