Apple Stole My Videos For Their AI

Apple Stole My Videos For Their AI

Hey everyone, I'm Sam Tucker from Samtime. Recently, I came across an alarming piece of information: Apple has allegedly stolen content from over 170,000 YouTube videos to train their new AI model. This is ironic because, while it's funny when I parody Apple content, it's certainly not okay for them to steal content from creators like me.

Apparently, because Apple brands itself as a "privacy company," they avoid using customer data for training their AI model. This is unlike Google or Facebook, who have fewer qualms in this area. Despite Apple's focus on privacy, they still need data to train their AI. So, where did they turn? The auto-generated subtitles from YouTube videos.

Auto-generated subtitles are an interesting choice, but they may not be very accurate. YouTube's transcription system often mistakes my English for Korean, for instance. Moreover, the insights they capture might not do wonders for Apple's image. Imagine an Apple AI bot telling users that the new iPad Pro bends easily or that new iOS features are just Android features repackaged.

This giant data grab wasn't even supposed to reach Apple's hands. It turns out the subtitle data was compiled by a nonprofit organization to help smaller companies, not trillion-dollar giants like Apple who can very well afford premium services. Apple isn't the only offender here either; Nvidia has been doing it too. And really, does a graphics card company need my content for their AI?

What might happen next? Well, my dad is somewhat of a YouTube creator himself. If Apple's AI trained on his content, their technology might become even less useful.

In the end, Apple clarified that they've only used this data for their open-source AI model. But this leaves the question: where is their real AI data coming from? Stay tuned; this case is not closed yet.

Keywords

• Apple
• AI Model
• YouTube Videos
• Auto-generated Subtitles
• Nonprofit Organization
• Privacy
• Data Theft
• Nvidia

FAQ

Q: What did Apple allegedly steal to train their AI model? A: Apple allegedly stole content from over 170,000 YouTube videos, specifically the auto-generated subtitles.

Q: Why did Apple use YouTube subtitles for training their AI? A: Apple used YouTube subtitles because, as a privacy company, they prefer not to use customer data. Instead, they turned to publicly available auto-generated subtitles.

Q: Are YouTube auto-generated subtitles accurate? A: Not always. YouTube's transcription system often makes errors and can even mistake languages.

Q: Was this subtitle data intended for large corporations like Apple? A: No, it was compiled by a nonprofit organization to help smaller companies train their AI models.

Q: Is Apple the only company accused of this? A: No, Nvidia has also been accused of using YouTube content in a similar manner.

Q: What was Apple's response to these allegations? A: Apple stated that the data was used only for their open-source AI model, though questions remain about where their primary AI training data comes from.