➡️ Exact delivery dates may vary, so some features may not be immediately available on the publication date. For your specific access dates, please contact your LivePerson account team.
🚨The timing and scope of all changes remain at the sole discretion of LivePerson and are subject to change.
Introducing Syntrix - Where AI promise meets actual proof
Welcome to the future of AI agent evaluation and live agent training. We are thrilled to introduce Syntrix, your new essential toolkit designed to optimize both AI and live agents.
In a world where "hoping for the best" isn't a strategy, Syntrix offers a simulation and training studio—a revolutionary AI evaluation and live agent training environment designed to sharpen your most important assets: your agents, both AI and live.
Navigating the complexities of modern CX requires more than just caution; it demands a dedicated space for experimentation. Since testing in production is never a safe option, Syntrix serves as your predictive evaluation layer and sandbox. By replacing manual, high-risk testing with automated, large-scale simulations, Syntrix allows you to stress-test policies and guardrail reliability before they ever reach a customer—delivering AI readiness that unlocks your ability to use customer-facing AI in production.
✨ Key capabilities
Hyper-realistic synthetic customers
Forget static scripts. Syntrix utilizes Generative AI to create synthetic customers that mirror real-world emotional ranges and complex customer journeys.
- Realistic chaos: Test how agents handle high-pressure interactions and "diverse personalities."
- Streamlined authoring: Quickly build archetypes and scenarios to reuse across your training modules.
Orchestration and bulk validation
Why test one-by-one when you can test many at once?
- Validation at scale: Launch parallel simulated conversations to continuously evaluate agent performance across a myriad of permutations.
- Targeted skill assignments: Instantly map specific scenarios to relevant agent skill sets to ensure specialized readiness.
The intelligence loop
Syntrix isn't just a playground; it’s a feedback engine.
- Adaptive coaching and tuning recommendations: Get immediate, context-aware feedback about live and AI agents on nuances like empathy and compliance.
💡 Core use cases
Quality assurance and optimization of AI agents
Replace labor-intensive testing with scalable solutions built for modern AI architectures.
- Accelerated optimization: Test and validate end-to-end AI agent performance, including handoffs to other agents.
- High-fidelity testing: Eliminate manual bottlenecks by generating multi-turn conversations at scale to ensure your AI agents are tested, evaluated, verified, and validated before ever speaking to a customer.
Live agent enablement and training
Accelerate an agent's time-to-proficiency for the most complex, emotionally charged conversations.
- Risk-free roleplay: Let agents practice against edge cases and non-compliant paths using synthetic data in a safe environment.
- Certification paths: Build structured onboarding where new hires progress through AI-driven scenarios to verify readiness before going live.
- Targeted coaching: Deliver objective feedback on empathy and brand compliance based on simulation outcomes.
👥 Who is this for?
- Training and QA teams: To design simulations and automate quality oversight.
- Technical leads: AI designers and bot managers looking to validate logic at scale.
- Frontline staff: Live agents practicing in a safe, risk-free environment.
Ready to start simulating?
Get started today with Syntrix AI agent evaluation and live agent training. It’s time to turn potential errors into immediate, actionable lessons within a safe testing environment. Contact your LivePerson representative to get started.
Harmony's GA release - Modernize your web messaging experience
We are excited to announce the General Access release of the Web Theme: Harmony, a modernized, "future-proof" web messaging theme that replaces the legacy "Ocean" and "Horizon" themes.
Harmony is a mandatory technical prerequisite for upcoming features, including Threaded Conversations.
The big picture
- Modernized UX: Creates a sleeker, more intuitive consumer experience.
- Future-Proofing: Serves as the technical prerequisite for upcoming platform innovations.
- Improved Performance: The underlying tech stack has been modernized, better leveraging our CDN for improved load time performance.
Feature area | Legacy themes (Ocean/Horizon) | Harmony theme |
|---|
Layout | Wider/classic layout | Vertical "Tall" layout (390w x 700h) |
Message bubbles | Distinct "speech bubble" tails on every line | Unified message bubbles (no more tails on every line) |
Menu UX | Separate emoji and attachment menus | Merged Emoji/Attachment menu |
Design | Older transcript and system message design | Updated transcript design (margins, font, positioning) and a new look for system messages |
Technology | Older underlying tech, greater dependency on backend fetches | Modernized stack thatdelivers content from servers closer to you, ensuring the app stays fast and stable no matter where you are |
Enhanced header | Optional/separate feature | Enabled by default when Harmony is used |
Enablement and configuration
Enablement
Harmony is available to all brands in the Conversation Cloud (CCUI) and does not require a support ticket to enable.
- New brands: The Harmony theme is enabled by default.
- Existing brands on a dedicated CCUI phase: Contact your LivePerson representative to enable the new Harmony theme. Once enabled, see below for more.
- All existing brands: An existing engagement can't be updated to Harmony. You must switch to a Harmony engagement and then apply any special configuration to that new engagement instance. The process below describes how to create the new Harmony engagement.
Add a Harmony engagement instance
- Log into Conversation Cloud.
- Navigate to Engage > Campaign Builder > Select Campaign (or create a new one).
- In the Engagement Window section, hover over the theme, and click Edit.
- Locate Harmony under the "System Themes" row.
- Click Duplicate Theme (System themes are read-only). The custom version will appear in MY THEMES.
- Click Next to validate/change Entry Point, and Next again to validate/change Behavioral Targeting.
- Click Publish.
Customization
Once the Harmony theme is duplicated, it can be customized via the 3-dot menu in the Engagement Window Library. The customization interface looks similar to the legacy editor but applies to the Harmony theme structure.
Known issues and considerations
- Custom CSS risk: Existing "on-page" custom CSS that worked with the Ocean or Horizon theme might conflict and break Harmony due to the new Document Object Model (DOM) structure. Brands must audit and test their site CSS prior to going live.
- System themes are read only: You can't edit the "Harmony" system theme directly; you must duplicate it first to create an editable custom version.
- Mobile SDK availability: Harmony for Mobile SDK is not currently available and is not prioritized on the 2026 roadmap.
Generative AI applications - Enhancing AI performance with Google Gemini
At LivePerson, we are committed to providing you with the most advanced and reliable Generative AI capabilities. To ensure we continue leading the industry in innovation and performance, we are pleased to announce the migration of our AI-powered applications to Google’s latest Gemini models.
This transition is driven by our ongoing partnership with Google, which grants us early access to cutting-edge models. Our extensive benchmarking has shown that Gemini models consistently perform as well as—and in several key use cases, better than—our previous models. By moving to Gemini, we are not only ensuring the long-term stability of our services ahead of upcoming model retirements but also delivering faster, more accurate, and more efficient AI experiences to help you better serve your customers.
➡️ All the rigorous safeguards you rely on today will remain in equal or improved standing with Gemini. Specifically:
- Zero data retention: Your data is processed in-memory and is never stored by the model provider.
- No model training: Google is strictly prohibited from using your data to train their foundation models.
- Robust guardrails: All of our existing Generative AI security guardrails are fully supported to ensure sensitive data is protected.
For a comprehensive overview of our security architecture and data handling policies, we invite you to review our updated Generative AI Security & Privacy Whitepaper.
Impacted products
The migration to Gemini affects the following LivePerson products:
- Automated Conversation Summaries
- Copilot Rewrite
- KnowledgeAI integrations in Conversation Assist
- KnowledgeAI agents built in Conversation Builder
- Routing AI agents built in Conversation Builder
- AI agents built in aiStudio
⚠️ Action required
All brands:
Review the Enhancing AI performance with Google Gemini article in our Community Center. This article has complete info on the process, the timelines, and product-specific guidance.
Brands using specific products:
If you use any of the following products:
- Automated Conversation Summaries
- Copilot Rewrite (NA and EMEA regions only)
- KnowledgeAI integrations in Conversation Assist (NA and EMEA regions only)
- KnowledgeAI agents built in Conversation Builder (NA and EMEA regions only)
New system prompts that are optimized for Gemini are available in the system, so preview of these prompts is now available. Also, please expect to find optimized custom prompts available in the system very soon. (We'll add them to your account for your explicit, manual selection for testing purposes.)
Soon the Gemini-optimized system prompts will become the default prompts. If your Production solution uses a default prompt, not a custom one, in effect you are switched automatically to Gemini at this cutover point.
At a later point, we will also be automatically cutting over all custom prompts to Gemini. This will impact you if you do not manually migrate your custom prompt to Gemini beforehand.
➡️ If you use any other impacted region or product, stay tuned, as we're still refining our migration schedules. We'll announce your calls to action via a release note.
New in the RCS channel - One-tap calendar scheduling and location requests
We are expanding the interactive capabilities of our RCS channel by introducing two new Suggested Actions:
- Schedule calendar event
- Request location
These actions allow you to move beyond simple text replies and provide functional, "one-tap" utilities that integrate directly with a user’s native device applications.
Schedule calendar event
This action streamlines appointments and bookings by allowing the RCS agent to trigger a pre-filled event in the user's native calendar app.
- Seamless integration: When the user taps this suggestion, their default calendar app opens automatically with the event details already populated.
- Customizable details: You can specify the event title (up to 100 characters), description (up to 500 characters), and start/end times.
- Example use case: A user receives an invite to visit a bank branch; tapping "Schedule a Meeting" automatically creates a calendar event for their upcoming appointment.
Request location
This feature introduces a secure and seamless way for you to ask for a user's current GPS coordinates.
- How it works: Instead of the user manually typing an address or searching for a pin, the brand sends a "Request Location" suggestion. When the user taps it, they are prompted to share their current location with a single tap.
- Key distinction: This is a pull mechanism. While "Share/View Location" allows the brand to show where they are, Request Location allows the user to send their location to the brand.
- Example use case: "Find the nearest service center" or "Track my delivery driver." The user taps the request, and their coordinates are sent directly to the LivePerson agent/bot to calculate the closest proximity.
To get started with suggested actions, see the complete JSON template, code examples, and full list of properties in the official documentation in our Developer Center.