EP 256: Microsoft's VASA-1 AI Deepfake: So good it's dangerous?

Is this AI tool too good to be safe? Maybe.  We'll be taking a look at Microsoft's impressive VASA-1 deepfake cloning technology. We'll answer the important questions:↳ What is this technology? ↳ Why do we need it? ↳ Is it so good that it's dangerous? Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion: Ask Jordan questions on Microsoft VASA-1Related Episodes:Ep 211: OpenAI’s Sora – The larger impact that no one’s talking aboutEp 157: Future of AI Video – Pika Labs 1.0, Runway updates, Meta Emu and moreUpcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [email protected] with Jordan on LinkedInTimestamps:01:20 Daily AI news05:15 Is VASA-1 dangerous?09:12 Potential dangers of AI outweigh potential benefits.12:14 Voice inflection changes, impressive text to speech.15:29 Generative content creation: impressive, concerning, and exciting.18:54 Quickly scalable technology opens exciting possibilities but troubling.21:23 Real-time face swapping with surprising computing power.24:35 Realistic motion control, capturing human-like tendencies.29:26 Use digital avatars for efficient training purposes.30:16 Microsoft research paper VASA 1 creates human-like avatars.34:20 Discussing generative AI's potential benefits versus risks.Topics Covered in This Episode:1. Explanation of VASA-1's capabilities2. Concerns around Microsoft's VASA-13. Discussion on deepfakes4. Potential Risks and Ethical Considerations5. Potential Benefits and Positive Use CasesKeywords:Microsoft's VASA 1, deep fake model, Everyday AI, livestream podcast, AI news, Adobe, Photoshop, Google, DeepMind, Microsoft PHY 3 mini model, OpenAI's SoRUP, AI deepfakes, talking head images, Jordan Wilson, misinformation, disinformation, generative AI, AI's impact on jobs, lifelike talking faces, digital twin technology, AI avatar, AI-generated speech, Mona Lisa, Baidu's Emo, Synthesia, HeyGen, Hour 1, DID platform, training, personalized learningSend Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info)