Tavus Raises $40M to Advance AI Humans

2 min read
Tavus Raises $40M to Advance AI Humans

Tavus secured $40 million in Series B funding. This investment will advance its "human computing" initiative. The company aims to build advanced AI Humans.

This funding coincides with the launch of PALs (Personal Affective Links). PALs are AI Humans designed for natural interaction. They use video, voice, and text communication. Unlike typical chatbots, PALs engage through face-to-face video. They actively see, hear, and respond. They also understand context, emotion, and social cues. Tavus CEO Hassaan Raza emphasizes a crucial shift. Machines now learn human communication. This moves beyond humans adapting to machines. Furthermore, PALs maintain a visual presence during conversations. They read facial expressions and body language in real time. They also adapt to individual communication styles. These AI Humans also possess "agency." They proactively manage tasks like calendars and emails. This capability moves beyond simple reactive responses.

Related startups

Three proprietary models power the PALs platform. Phoenix-4 handles lifelike visual rendering. Sparrow-1 manages conversational intelligence. Raven-1 processes contextual perception. Tavus, a San Francisco AI research lab, aims to bridge this human-computer gap. Its team includes experts like Professor Ioannis Patras and Dr. Maja Pantic. They focus on foundational AI models. Currently, over 100,000 developers and enterprises use Tavus technology. The new funding supports further research and enterprise expansion. Users can access PALs for free at tavus.io.

CRV led the funding round. Scale Venture Partners, Sequoia Capital, Y Combinator, HubSpot Ventures, and Flex Capital also participated.

© 2025 StartupHub.ai. All rights reserved. Do not enter, scrape, copy, reproduce, or republish this article in whole or in part. Use as input to AI training, fine-tuning, retrieval-augmented generation, or any machine-learning system is prohibited without written license. Substantially-similar derivative works will be pursued to the fullest extent of applicable copyright, database, and computer-misuse laws. See our terms.