Microsoft has unveiled a new feature for its Copilot assistant designed to make conversations feel more natural. The experimental “Portraits” option, now available in the US, UK and Canada, gives users a choice of 40 animated human avatars that display facial expressions during real-time voice chats.
The feature was developed in response to feedback from Copilot users who said they would feel more comfortable talking to a face rather than just a voice. Announcing the update on X, Microsoft AI chief executive Mustafa Suleyman said Portraits was built to make verbal interactions with Copilot more approachable. Users can select an avatar and pair it with a voice, aiming to make conversations flow more naturally, like real human dialogue, compared to text-based exchanges.
You can now talk to a Copilot Portrait in real-time! We’ve heard from some users that they’d feel more comfortable talking to a face when using voice. So in the US, UK + Canada, we’re rolling out a new Copilot Labs experiment where you can talk with animated portraits. pic.twitter.com/KEubl1Kv1N
— Mustafa Suleyman (@mustafasuleyman) September 29, 2025
The development follows Microsoft’s earlier experiment with “Copilot Appearances” in July, which used cartoon-style characters. The new Portraits option, however, is based on stylised human images rather than animated blobs, bringing the assistant closer to a lifelike personality.
Powered by Microsoft’s VASA-1 technology
According to a post on the Copilot Discord server, the Portraits experiment uses VASA-1, a technology created by Microsoft Research. The system generates facial expressions, head movements and accurate lip-sync from a single image, removing the need for complex 3D modelling.
“This experiment leverages VASA-1, an advanced AI technology developed by Microsoft Research to create real-time, visual AI conversations,” Microsoft explained. “The innovation allows us to generate natural facial expressions, head movements, and lip-sync from just a single image — no complex 3D modelling required.”
By combining visual cues with audio responses, Microsoft hopes to bridge the gap between traditional chatbots and human-like interactions. The company says Portraits is part of its ongoing work to make Copilot more engaging, while keeping safety and reliability at the forefront.
Careful rollout and safety measures
The launch of Portraits comes at a time when rival AI platforms are exploring their own avatar-based assistants. X recently introduced 3D avatars for its Grok chatbot, including an anime-inspired companion that has sparked controversy for offering an NSFW mode. Microsoft appears to be taking a more cautious approach, limiting access to Portraits to address potential concerns about harmful interactions.
In a statement on the Copilot blog, Microsoft said: “Each portrait is intentionally stylised and not photorealistic, offering you visual variety while offering the same trusted Copilot intelligence and safety features you rely on.”
Availability is currently restricted to a limited number of users aged 18 and older. Microsoft has also set daily and session-based time limits to help manage usage, alongside clear indicators that remind people they are interacting with artificial intelligence.
The Portraits experiment reflects Microsoft’s effort to make Copilot more approachable, while ensuring safety and transparency remain in place as the technology develops.
