Summary

Every other app and service now seems to be tacking on AI functionality to attempt to entice weary users or satisfy shareholders. Sometimes, AI features work well. Often, it just results in junk output (at least for now) and a privacy nightmare. Here’s how Linux can help you escape the AI apocalypse.

Everyone Is Adding AI to Their Products and Services

I’m no curmudgeon: Generative AI is an awesome new technology, and I think that anyone with the hardware should try hosting their ownAI chat botsandimage generatorsto learn how they work and prepare for the future. It’s the haphazard application of AI, without any thought for users, and that makes products worse that I object to. For example:

This effort to add AI to everything isn’t stopping, despite users' trepidation. Windows, MacOS, and iOS are all plowing ahead with new AI features, and at no cost to you. This raises the question of how they plan on profiting from it (hint, the answer is probably by monetizing the data you feed to them).

The Kubuntu Focus Ir16 Gen 2 laptop open on a desk in front of speakers and a headset.

For AI to Work, It Needs to Scrape Your Data

Generative AI isn’t pushing the limits of creativity, it’s just an algorithm that mashes together pieces of data it’s been trained on. That means generative AI needs a constant stream of new information to work with to stay up-to-date.

Initially this was provided by scraping the public internet (arguably stealing real people’s content in the process). But that’s no longer enough, and each AI vendor wants their own unique data that puts them above the competition.And they want it from you.

When you interact with an AI, it records your input. This isn’t so much of a problem when you’re just firing questions at a chatbot for a laugh, but it could be a huge problem when the AI feature is baked into your operating system. If you aren’t careful about what you agree to when an app or system update is installed, or you buy a new device, your activity is potentially being recorded for AI training.

This doesn’t just make AI output unoriginal, It’s a massive privacy issue: how can you be sure some details of your private life that have been submitted to an AI platform won’t be used to generate someone else’s short story, business email, to create images, or justrevealed to the public outright?

Take Control Back By Switching to Linux

Alright, I’ve kvetched about AI privacy enough, so what can you actually do about it? The answer is pretty short.

Take control of your computing experience by switching to Linux, the 100% free, open-source operating system. Major Linux distributions likeUbuntu and Fedoradon’t have any AI baked in, and if they do in the future, you’ll be able to just uninstall it or switch to adifferent Linux distribution. If you do want to use AI-enabled tools, you canchooseto, not beforcedto.

The open-source community that develops Linux applications is notoriously privacy conscious, so if your favorite app decides to force new intrusive features on you, you’re able to usually find a safe alternative (for example,LibreOfficeis a totally free alternative to Microsoft Office, without online services and AI integrations).

Privacy is more difficult on mobile devices, as there aren’t any widely available open-source mobile phones and tablets. But at least your PC can be your own, private, AI free oasis.