Microsoft's new AI is the next revolution of the Internet

Microsoft's new artificial intelligence can instantly generate realistic video of a human's face while speaking.


VASA-1 is Microsoft's new image-to-video artificial intelligence that blurs the line between real and fake video, and can generate realistic video with real lip movements from an image or audio file

Microsoft notes that VASA-1 can not only handle lip syncing with audio playback, but also capture a range of facial expressions and subtleties and natural head movements that contribute to perceived realism.

Microsoft's new AI can produce 512 x 512 pixel video at 45 frames per second, and it is capable of real-time video of a person's face speaking at 40 frames per second.In addition to the useful uses of VASA-1, its potential for abuse is high, raising concerns about creating highly realistic deepfakes.


Responding to the concerns, Microsoft says that VASA-1 is primarily a research project. The company acknowledged the inherent risks, stating, "Like other related content generation techniques, it can still be exploited to impersonate humans." "We oppose any behavior to create misleading or harmful content from real people."Microsoft says it won't release its new AI to the public right now and is waiting for strong regulation to reassure users. Microsoft says that this new technology should become the norm for the entire industry.