Newsbite analysis · AI, Cloud · New AI Model
Google News
Introducing Gemini 3.8 Live with Live Avatar
Google has revealed Gemini 3.8 Live with Live Avatar, successfully integrating near real-time video generation into its AI dialogue models. This feature aims to provide a more dynamic and interactive experience for users, particularly in enterprise settings, allowing virtual agents to listen, speak, and visually engage with users seamlessly. The Live Avatar can adapt across 97 languages while maintaining synchronization in lip movements and expressions. Organizations can also personalize these avatars to fit their branding and enhance user interaction through tools and background task handling.
Act nowConfidence: HighImmediate horizonStory: First report
EvidenceOfficial announcementIndependent evidenceCompany only
Published 25 Sept 2026, 09:47 UTC3 min read
Lead image — Google News
Brief
NewsBite reading:Google launches Gemini 3.8 Live with enhanced Live Avatar features for enterprise interactions
TL;DR
- Launch of Gemini 3.8 Live with Live Avatar, enhancing digital interactions.
- Near real-time video and audio processing for dynamic engagement.
- Supports 97 languages with fluent lip-syncing and expressions.
- Enterprises can create custom avatars to match their brand.
- Strengthened trust and transparency features with AI content watermarks.
Interesting facts
- Gemini 3.8 Live with Live Avatar can perform background tasks during dialogue.
- It can switch languages mid-conversation without visual drift.
- Custom avatar creation is restricted to enterprise clients.