TLDRocket
Sign in

Multimodal Models

96 summarised stories about Multimodal Models, each linking back to the original source. Browse all topics →

+ Follow this topic

Friday, 2 October 2026

48% thought Tavus’s AI was human

tavus.io 1 week ago 43 ● 2 sources

Tavus introduced Griffin, a human interaction model meant to handle real-time face-to-face video conversations by listening while tracking expressions and pauses. In a one-minute video call test, 48% of participants thought they were speaking with a real human, compared with a max 2% pass rate for prior systems. Tavus says Griffin-Lite is available to a select group for testing and a wider release of a more powerful model will follow.

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.