Twilio Changelog

As previously communicated, Voice JS SDK 1.x will reach end of life April 1, 2025. After that date, calls using 1.x may fail to connect, causing disruption to any applications still using it. To avoid disruptions, please migrate to 2.x as soon as possible.

Introducing the Regulatory Compliance API Private Beta

We're launching the Regulatory Compliance API Private Beta, aimed at improving the Twilio compliance onboarding API experience.  This Private Beta is opt-in, and we're inviting interested customers to express their interest. 

Capabilities of the Regulatory Compliance API Private Beta include:

• Fewer Calls to Create Bundles: The new API (v3) simplifies the process of creating Regulatory Compliance Bundles by reducing the number of requests needed. Unlike the v2 API, which requires multiple steps per requirement (ex: Authorized Individual, a Business Address, and a Business Name), the new API consolidates these into a single request. With more complex registrations, the benefits are greater.

• Improved Error Codes: Errors are now surfaced in the API and error dictionary allowing for a more programmatic approach to error handling and less time troubleshooting. 

• Single Call to Fetch Bundle Details: Getting bundle details is simplified into a single request.

**Important Note:  Private Beta access is limited and the Twilio Product team will reach out to interested customers regarding access. More information is available here to learn more about Twilio’s general Beta Guides.

Beginning today, we’re reorganizing the Flex section of Twilio Console to make it easier for Flex administrators to find what they need. 

The Flex administrator experience in Console has grown significantly, and the existing navigation structure is no longer able to support all the content we’ve added. To address this issue, we’re redesigning the Flex navigation structure. We’ll be introducing changes gradually, one category at a time.

Today, we added the first new category: Users and access. This category contains the Single sign-on (SSO) and Automatic logout configuration pages. 

Soon, you’ll see more new Flex navigation categories. As existing pages are moved, if you try to access the old URLs, you’ll see redirect notices to explain where you can find the new pages. If you have bookmarks or other links to the old URLs, we recommend updating them to the new URLs when you see these notices. After a transition period, the old URLs will simply redirect to the new pages.

As of March 17, 2025, ConversationRelay can now support healthcare use cases that contain protected health information (PHI) for organizations that are subject to the Health Insurance Portability and Accountability Act (HIPAA). Twilio will sign Business Associate Addendums (BAA) with covered entities and business associates for HIPAA Eligible Services.

As of March 17, 2025, Flex Webchat 3.x.x. can now support healthcare use cases that contain protected health information (PHI) for organizations that are subject to the Health Insurance Portability and Accountability Act (HIPAA). Twilio will sign Business Associate Addendums (BAA) with covered entities and business associates for HIPAA Eligible Services.

To learn more about how to build on Twilio for HIPAA compliance, see Architecting for HIPAA on Twilio.

We’re excited to announce that Flex Webchat 3.x.x is now generally available (GA). Webchat 3.x.x is a chat widget that you can embed on your website. The widget helps your customers chat with an agent without leaving your webpage. It’s natively integrated with Flex UI, built on Flex Conversations, and hosted by Twilio.

Webchat 3.x.x is simple to set up: you create a deployment key, connect it to a messaging address, and then copy the Flex-generated embed code to your website. Compared to other Flex Webchat options, Webchat 3.x.x also includes enhanced security. You can use the out-of-the-box configuration or customize your Webchat to request user information and gather context data before the start of a chat.

Our latest version, Webchat 3.2.0, was released on March 10, 2025. For information about updating from Webchat 2.x, see our Webchat migration guide. 

To learn more and get started, see the Webchat 3.x.x overview.

Easily verify if you can port-in your desired US local phone numbers to bring Voice and Messaging capabilities from your current carrier to Twilio.

On March 18, 2025, SendGrid updated the information included in email headers to enable customers to utilize Gmail’s Feedback Loop. Previously SendGrid populated the header with x-feedback-id, which limited customer usage of gmail’s feedback id. Now senders will have better resources to identify spam complaints from individual campaigns.

Generative Custom Operators are a new class of Voice Intelligence Language Operators, backed by Generative AI, that support a broad range of sophisticated natural language analysis tasks with a high degree of accuracy. Generative Custom Operators are now available in public beta.

Generative Custom Operators allow you to provide natural language instructions describing what you want to understand from customer calls. These instructions are then fed to a Large Language Model (LLM) along with the call transcript to analyze the customer interaction and provide actionable results.

This feature significantly extends the flexibility, power, and accessibility of custom Language Operators to address a variety of conversational intelligence use cases. Examples include (but are not limited to):

• Call Scoring: Assign an overall score based on agent performance against customized criteria.
• Script Adherence: Evaluate whether agents stuck to the script and achieved their goals.
• Tailored Summaries: Generate call summaries in the format that best fits your business needs.
• Upsell Identification: Use the context of the call to identify high-likelihood upsell opportunities.
• Lead Qualification: Use buying signals to classify leads into categories like 'Hot' or 'Cold'. 

Support for Generative Custom Operators is now available both in the Voice Intelligence Console and via API. Advanced options are also available for JSON structured Operator results and providing training examples.

To start building with Generative Custom Operators for Voice Intelligence, including use case examples and best practices, please see our getting started guide.