New Microsoft AI tool makes still images talk (and rap, sadly)

For over 500 years, humanity has not had to put up with the Mona Lisa being animated to lip sync over a crudely bellowed novelty rap song. That streak of civilization has sadly been ended by serial tech botherer Microsoft, which has unveiled its latest artificially intelligent horror on planet Earth. VASA-1 is able to take a still image of a person and realistically animate it using nothing but spoken audio, producing lifelike video. The tool even works in real time. Microsoft said that the tool ‘paves the way for real-time engagements with lifelike avatars that emulate human conversational behaviors.’ But before you go dashing off to Snapchat to use this technology in definitely hilarious ways, know that Microsoft has decided to not release VASA-1 to the public, on the grounds that it might be used for the express purpose it was designed for – making people look like they’ve said things they haven’t. Microsoft are therefore keeping the power to do this for themselves, which gives us a warm and fuzzy feeling of safety and security inside (note: sarcasm).