Apple and Meta Fight About Privacy And the EU Weighs In

A few days ago we saw that talks between Apple and Meta had amounted to nothing. There was talk about including Meta’s AI efforts into Apple Intelligence, unti Apple said “Forget it” or somethign to that effect. I skimmed the headlines but don’t remember the details.

This is significant. The reason for this is simple. The EU is being criticised by Apple for the Digital Markets Act and Apple is trying to give the EU a bad name by saying “We won’t make Apple Intelligence available in the EU because we’re not clear about the DMA.

The reality is that Apple is tarring and feathering itself with this move. The key reason to upgrade to M-Laptops and Apple 15 Pro and Pro Max phones is Apple Intelligence, but if Apple Intelligence is not available in Europe then that unique selling point vanishes and the lure of new, expensive devices, fades away.

Months ago Facebook gave us the option of either paying to have privacy, or agreeing to be data mined. For a month or so I agreed to pay the fee, but then almost immediately changed my mind. Meta defends charging fee for privacy amid showdown with EU

Today, the European Commission (EC) announced preliminary findings that Meta’s so-called “pay or consent” or “pay or OK” model—which gives users a choice to either pay for access to its platforms or give consent to collect user data to target ads—is not compliant with the DMA.

When someone says “pay or be mined” this is not a choice. This is an ultimatum. I wouldn’t mind, if Facebook was small enough to be ignored but it isn’t. For many, many people Facebook has become the internet. It is a web portal with messaging, shopping, dating, groups events and more. It has a monopoly, especialy through the acquisition of Whatsapp and Instagram years ago.

I don’t agree to them data mining my information, and I don’t agree to the way they control my timelines with algorithms.

Apple Intelligence and Meta

That Apple is in talks with OpenAI and Google, but abandonned talks with Meta/Facebook speaks volumes, in my opinion, and it is just an opinion. I haven’t seen it documented. Facebook was recently told that it could not mine user data to train its AI models in Europe. Specifically “Meta said the Irish Data Protection Commission (DPC) asked the company to delay training its large language models on content that had been publicly posted to Facebook and Instagram profiles.” This would explain why Apple is not happy to deal with Meta for the forseeable future. If the EU has to defend Europeans in such manners, it indicates that Apple too, feels that it must “defend” its users, or at least it’s market position.

Microsoft’s AI Boss

Recently Microsoft’s AI boss said something that is deeply immoral, and wrong.

I think that with respect to content that’s already on the open web, the social contract of that content since the ‘90s has been that it is fair use. Anyone can copy it, recreate with it, reproduce with it. That has been “freeware,” if you like, that’s been the understanding.

That is wrong, Content on the web is freely available but unless specifically mentioned it is still copyrighted. Content belongs to the creator of that content. My blog and website are free to access but I still hold the intellectual property to what I have shared. Images have copyright notices.

Websites like flickr give us the option of using creative commons licences, but those licences, especially CCbyA and others are clear. It is free to share and link to certain images but for free, and non commercial use, with attribution.

In other words plenty of content that is being hoovered up by AI is being stolen which is why Creatives are fighting against the spread of AI. Two or three decades ago people were worried about napster, torrents and peer2peer networks. Even YouTube was in jeopardy over the issue of copyrighted content.

The difference, in the AI Age is that corporations are stealing, from everyone, by ingesting the web, and then spitting it back out, and charging people for the right to search for content, trained on the intellectual property that might have been stolen from them.

I am a Fan of AI and Machine Learning as a Media Asset Manager and Archivist

I am a fan of AI and Machine learning because as a Media Asset Manager I find that tools that transcribe, log and more, automatically are great. I love the idea that if I invest enough time in teaching the machine about locations, faces, objects, contexts and more, that AI will help produce a deeply researched media archive for future generations to explore and learn from.

I spent two years as a video archivist and Media Asset Manager so I see the value in a machine learning from us, how to sort video files, audio, speech and more. The work that I have done, if it can be ingested by machine learning, could then speed up, and make more accurate, the work that I did upstream.

The Big Picture

I believe that self-hosted machine learning, such as Apple Intelligence is trying to be, is fantastic. I saw projects like Personal AI and I love the concept. What I don’t like is that it’s built on top of ChatGPT which is not private. Morten discussed the value of small language models recently.

The point of small language models is that they are trained on domain specific content rather than everything. They are designed to be niche. ChatGPT, Gemini and Copilot are LLMs that require huge resources to work. Ollama, also requires huge resources to work.

He argued that we don’t need large language models, but that specialist models are the right approach and I agree with this.

OpenAI and Microsoft want to steal or missapropriate all of the world’s data, and the EU is telling them that they can’t. They need to respect people’s privacy. Private data should remain private. By private I mean personal posts on social networks, even if they’re public.

Conclusion

That Facebook and Apple have decided not to collaborate could imply that Meta wants too much access and freedom to use the data that they would siphon from our iDevice use. It consolidates that we should not be using FB, IG and other Facebook projects. I write Facebook, rather than Meta, because the company is the same. they just changed the name for PR reasons, rather than their ethos. If Facebook had no monopoly I would have dumped them, years ago.