• FryAI
  • Posts
  • Adobe unveils video upscaling model

Adobe unveils video upscaling model

Good morning, and happy Monday! Prepare for a feast of AI news that is fresh, crispy, and full of flavor. 🍟

(The mystery link can lead to ANYTHING AI related. Tools, memes, articles, videos, and more…)

Today’s Menu

Appetizer: Adobe unveils video upscaling model ☝️

Entrée: Apple to choose OpenAI over Google? 📲

Dessert: Baltimore HS athletic director fired over AI video 🎥

🔨 AI TOOLS OF THE DAY

🧥 Fashion Advisor AI: An AI fashion expert, able to solve all outfit-related questions. → check it out

👨‍🏫 AI Flash Cards: Create flash cards from any text or file. → check it out

😃 What If?: A face swapper that lets you become anyone you want. → check it out

ADOBE UNVEILS VIDEO UPSCALING MODEL ☝️

The baby in this new demo is ADOBE-able. 😃

What’s new? Adobe researchers have introduced a new generative AI model called VideoGigaGAN, which aims to upscale blurry videos at up to eight times their original resolution.

How does it work? VideoGigaGAN is designed to deliver more fine-grained details without introducing the “AI weirdness” often associated with generative AI models of this kind. Generative Adversarial Networks (GANs) have been effective at enhancing still images but face challenges when applied to video, often resulting in flickering and other unwanted artifacts. Other upscaling methods may avoid these issues but compromise on sharpness and detail. VideoGigaGAN seeks to combine the advantages of existing GAN models with minimal flickering and distortion across output frames. In Adobe’s demo clips, the model generates impressive skin textures and eyebrow details that maintain a natural appearance without flickers during motion.

What’s the point? Although VideoGigaGAN is currently a research preview, it shows how Adobe continues to revolutionize photo and video editing with AI. In this way, Adobe continues to lead the charge for more sophisticated upscaling technology in future photo and video editing software.

APPLE TO CHOOSE OPENAI OVER GOOGLE? 📲

Q: How do you make an apple turnover?

A: Roll it down a hill. 🍎

What’s new? Apple has been in talks with OpenAI to power its next iPhone operating system, iOS 18.

What’s the significance? Previously, reports indicated Apple was considering a partnership with Google’s Gemini chatbot for new iPhone features. This new report suggests that Apple either had a change of heart or is considering a way to incorporate both features from Google and OpenAI into iOS 18, which is set to release later this year. Whoever wins this bid to power Apple’s devices will be in a great position and seemingly drive the push forward for on-device AI. Neither Apple nor OpenAI has commented on the matter, leaving the specifics of the potential collaboration uncertain.

BALTIMORE HS ATHLETIC DIRECTOR FIRED OVER AI VIDEO 🎥

I got fired from a keyboard factory. They said I wasn’t putting in enough “shifts.” ⌨️

What happened? A high school athletic director in a Baltimore suburb was arrested for allegedly using AI to impersonate the school’s principal, spreading racist and antisemitic hate speech as part of a revenge plot

What’s the story? The disturbing audio surfaced in January and seemed to feature Pikesville High School Principal Eric Eiswert making offensive comments about students of color and Jewish teachers. However, Eiswert denied the allegations and pointed to Darien as the likely culprit. Further investigation revealed that Darien allegedly used AI technology to create the fake recording as retaliation against Eiswert, who had launched an inquiry into Darien’s potential mishandling of school funds. The police ultimately tied Darien to the AI-generated recording through an email account under an alias. Darien now faces charges of theft, stalking, and retaliation against a witness.

What’s the significance? This story outlines another case of how AI-generated content is sometimes being used for malicious purposes. In this case, it led to an arrest. It’s worth noting, however, that the investigators were unable to figure out who was behind the video until they found an email attached to his name. This goes to show how difficult it is to trace creators of malicious AI-generated content, highlighting the difficulty of enforcing punishment for distributors of harmful deepfakes.

MONEY MONDAY 🤑

People are discovering innovative (and sometimes wacky) ways to make money using AI. Check out today’s featured video:

HAS AI REACHED SINGULARITY? CHECK OUT THE FRY METER BELOW:

The Singularity Meter Rises 1.0%: Teacher uses AI to make boss racist

What do ya think of this latest newsletter?

Login or Subscribe to participate in polls.