© AFP
The Mona Lisa can now do more than just smile thanks to Microsoft’s new artificial intelligence. But making the AI tool immediately available to the general public is not yet an option.
VASA-1, Microsoft’s new AI model, can create video images of the people in the photo based on still images of faces and audio fragments. They then get natural facial and head movements and the lips move as if they were really speaking.
Works of art, such as the Mona Lisa, can also be brought to life in this way. In a demonstration video, Microsoft researchers showed how they animated the Mona Lisa. For example, you can see how Leonardo Da Vinci’s artwork suddenly starts rapping based on an audio fragment by actress Anne Hathaway.
READ ALSO. Podcast company from Hasselt translates Dutch speakers into French in their own voice: “Sounds lifelike”
In other examples, where no art but real photos were used, it is sometimes barely noticeable that it is AI. That is why Microsoft is not immediately making the VASA-1 model available to the general public. The tool can easily be misused to spread misinformation or further disrupt the creative industries.
(Continue reading below the video)
This contains inserted content from a social media network that wants to write or read cookies. You have not given permission for this.
Click here to adjust your preferences
And so Microsoft works like OpenAI. Open AI already introduced Sora, an AI-generated video tool, in February. But Sora has so far only been made available to professional users and cybersecurity professors for testing purposes.
“We oppose any conduct to create deceptive or harmful content from real people,” Microsoft researchers said in a blog post. “The company has “no plans to” publicly release the product until we are confident the technology will be used responsibly and in accordance with appropriate regulations.”