Apple Anthropic Nvidia Stole YouTube Data to train AI Models | mkbhd

Introduction

What's Shake and Tech Fam, Today's News is Juicy!

Let's dig in. Apple, Anthropic, and Nvidia are in hot waters for training their AI models using transcripts from over 170,000 YouTube videos. This trove included work of big internet stars and clips from major television shows. The concern? These transcripts were scraped without the original creator's permission and may contain copyrighted material.

Meanwhile, record companies are suing Verizon, claiming they profited from people pirating music. Uh-oh, someone's not paying their royalties.

Anyways, Meta is letting researchers dig into Instagram's data to see how it affects young users. Good move or creepy? What are your thoughts?

Next, TikTok lost its recent legal challenge to the EU's decision to classify it as a very large online platform.

Google has announced Project Oscar, an open-source platform allowing developers to create AI agents to help manage their projects. Surprisingly, Apple released a new ad for Safari that continues to beat the dead horse by saying they care about privacy.

Last but not the least, the FTC has banned the health app Kaya, which aimed to be an AI pecker protector that could detect 10 STIs.

That's all for today, tech fam! Subscribe for more.