Twilio Conversations | Nov. 17, 2026

Twilio Conversations now supports AI agent supervision

You can now use Twilio Conversations to supervise your AI agents. This public beta adds two Conversation Intelligence operators that check whether your AI agents resolve customer requests and give accurate information, whichever model or vendor powers them. Conversation Insights shows the results across all your conversations.

This release includes:

  • Task Completion operator: Labels each conversation as resolved, unresolved, escalated, or declined by the agent, and gives a one-sentence reason. It only evaluates the first task.
  • Hallucination Detection operator: Checks the agent’s statements against your Enterprise Knowledge and the conversation itself, and returns the ones it flags. Claims it can’t check are listed as context gaps.

  • Conversation Insights support: Aggregate the results to spot recurring patterns and track quality over time. You can also send them to a data warehouse, such as Snowflake or Databricks, with Twilio Destinations.

Both operators run at the end of any conversation routed through Conversation Orchestrator. They use a large language model as a judge to evaluate the interaction. 

Read the blog post for the full story, or get started with our docs.

 

Conversation Intelligence Beta

Additional Resources

Blog

Read more about our latest product updates, product tutorials, and community projects.


Docs

See API reference documentation, quickstarts, SDKs, and multi-language code samples.

Events

Find upcoming events and join us virtually or in person to learn more about our products.