Have We Unleashed a Monster? The Threat of Deepfakes

Welcome, humans.🎉 We have something special for you! Skip the reading and experience the future of content – watch our AI-generated video with a realistic avatar narrating the post for you. Move over Hollywood, there's a new master of illusion – meet VASA-1, Microsoft's game-changing AI video tool. And finally, watch out! Our monthly update on AI's impact on work is ready for you to explore. Cache Cookies delivers your weekly dose of groundbreaking tech news, cool apps, brain-boosting knowledge, the latest on AI, and the lowdown on AI's wild side.

What You Need to Know About AI this week

• 🎉 Special Treat: Experience AI-Generated Video with Avatars

• VASA-1: Unveiling the Deepfake Threat ⚠️ 🎭

• 👀 Watch Out! AI's Impact on Work Update is Here

FROM OUR PARTNERS

🎉 We have a special treat for you! AI-generated video with an avatar narrating the post for you

Experience the future of content - read the post Have We Opened AI's Pandora's Box of Deepfakes? or watch our AI-generated video created with Synthesia with a realistic avatar. Instead of reading just click on the YouTube Video below!

Yes, this video was made with Synthesia, the Nvidia-backed UK AI tool! Imagine transforming your corporate videos, reports, and presentations with the same captivating ease. Choose from 160+ diverse avatars, 130+ realistic voiceovers, and customize in over 120 languages!

Next newsletter, we'll show you how! Get ready for a step-by-step guide to creating a professional avatar presentation in under 30 minutes.

Ready to take your videos to the next level? Try Synthesia today!

Have We Opened AI's Pandora's Box of Deepfakes?

Remember Sora, that mind-blowing AI video generator from OpenAI? The one that could create clips so realistic they could rival Hollywood special effects? Well, buckle up, because Microsoft just dropped VASA-1, and this new AI tool is pushing the boundaries of video creation even further.

VASA-1 takes things to a whole other level. Feed it a single photo and an audio track, and it'll transform them into a fully animated, synchronized video, like the one of raping Mona Lisa.

But here's the kicker: both VASA-1 and Sora, despite their jaw-dropping capabilities, aren't available for public use yet. The developers are holding them close, like shiny toys they don't quite want to share. Is it because they've stumbled upon powerful technology that could be easily misused – a digital Pandora's Box?

Why Does It Matter to Me 🦎 ?

VASA-1 isn't just about creating entertaining videos; it has the potential to revolutionize fields from education to doing business in corporations. Imagine history lessons where students can interact with lifelike avatars of historical figures, bringing the past to life in a way textbooks never could. Think of how nurse avatars could enhance telemedicine and improve patient care. Or consider personal trainer avatars guiding fitness enthusiasts through vigorous workouts on tall LED screens. Even loneliness could be combated by virtual companions offering support and connection. And this isn't some distant future – companies like Heineken, BSH, and Xerox are already using similar technology from Synthesia AI to streamline communication, training, and customer service.

Where are the Cache 🦎 Cookies 🍪 ?

While the potential benefits are exciting, it's crucial to ask ourselves why technologies like VASA-1 are being developed and the very real dangers they present. Imagine the chaos if anyone could create convincing deepfakes of public figures, putting words in their mouths that they never uttered. In an election year like 2024, with coming major elections in some of the world's most populous nations such as United States, Brazil and India the threat of political manipulation is immense.

Even though Microsoft has wisely withheld VASA-1 from public use, the technology behind it is readily accessible. Open-source models like Stable Diffusion allow individuals to build their own deepfake applications (here is a simple step-by-step procedure on how to create a deepfake 😲🤯). According to Divyendra Singh Jadoun, AKA 'The Indian Deepfaker' creating deepfakes has become easier, faster, and cheaper. A small team in India took just three days and 600,000 rupees (about $7,200) to create and edit a political election relevant deepfake using text-to-video systems in a workshop.

The process involves training AI models with a politician's voice and creating their digital avatar, which takes about a week. After training, personalized videos can be quickly generated, with voice cloning requiring only a few minutes to produce. This capability for microtargeting — allowing a politician to create individualized messages for a large audience — has never been available at such a scale. It raises concerns about the potential use of deepfakes in political campaigns and the broader implications for elections and public opinion.We've already seen the damage AI-generated disinformation can cause, from faked audio recordings influencing elections in Slovakia to attempts to discourage voting in New Hampshire. Now, imagine the impact of realistic deepfake videos. If you're still skeptical, watch BuzzFeed's chillingly realistic Barack Obama deepfake – a stark reminder of what's possible.

Are there any ways we can protect ourselves and our children?

The fight against deepfakes is complex. In Europe, the newly accepted EU AI Act mandates transparency. Creators must disclose the origin of their deepfakes, including the techniques used. This could involve watermarking. Meanwhile, the US is taking a legal approach. The Federal Trade Commission's (FTC) new rule prohibits impersonating government and business entities (with plans to expand to individuals).

But regulations and legal recourse are just part of the equation. We can also empower ourselves. By staying informed about these measures and developing a critical eye toward online content, we can be more discerning. Additionally, AI tools like Sentinel, the WeVerify Chrome plugin, and Microsoft’s Video Authenticator can be used to test suspicious images, audio, or videos. While these tools aren't foolproof, they offer another layer of defense.

We're committed to exploring solutions. Would you be interested in a review of these deepfake detection tools?

Cache Cookies Monthly Look at AI's Impact on Work !

Brace yourself! According to Goldman Sachs, by 2030, AI could wipe out a staggering 300 million jobs globally. Up to 14% of us may need to completely reinvent our careers. In Western nations, two-thirds of jobs face the threat of automation, and a quarter could soon be in the hands of AI. The future of work is closer than you think.

The AI-Powered Workplace: From Chatbots to CEO-Bots, the Future of Work is Here

Imagine this: A rough day at work? Pi.ai is your always-available, understanding confidante. Swamped with new hires? Synthesia handles the onboarding for you. Feeling hangry? Waymo's got your back with fast food delivery. And get this – dissatisfied with your CEO? Time to talk to Olivia, your HR manager, about some fresh options – maybe even Mika, the cutting-edge humanoid robot!

NetDragon Websoft made a groundbreaking move by appointing an AI, Tang Yu, as CEO. This sent shockwaves of excitement through employees and shareholders, causing their stock to skyrocket.

Imagine the cost savings if more companies ditch high-paid CEOs for AI! A tireless, unbiased AI leader, available 24/7, could revolutionize compensation – paying you directly for your work. ChatGPT has proven the power of AI in the workplace. The tools exist now to save time, unlock knowledge, boost productivity, and create employee-centric workplaces. The sky's the limit – AI can accelerate business growth and drive a culture of innovation. But at what cost?

We are witnessing a profound transformation of our roles and functions in the world of work, and it is our duty to stay informed. Follow Cache Cookies to anticipate these changes!

Job opportunities

That's it for now! Want more tasty AI insights? Visit our website and connect with us on LinkedIn @Mila and @Zeynep. While you're there, consider spreading the word – share Cache Cookies with all your friends and colleagues who also like to stay sharp!

We love feedback! Leave your comments.