[untitled]
5 comments
This whole thing is just one big advertisement and generally suggest nearly pointless things.
The ecological impact of AI for me is one of the biggest reasons to use it sparingly, even though I like technology. It‘s not as if we weren‘t in dire straights before its advent, but with all the additional datacenters being built and additional power and water required, we are just accelerating down the slope. Then come the ethical implications. And most of it, just to make more profit… profit for the few > well being of the many (now & for future generations)
You can’t fully “turn off” AI, but you can limit how much of your data fuels it. Even small actions matter, they signal that consent and transparency still count. The problem isn’t AI itself, it’s the incentive to harvest everything for profit.
> you can choose to pause AI by adjusting privacy settings, disabling AI features in apps, or requesting data removal from companies.
This is mostly futile regardless if it makes you feel better.
LLMs have been trained on everything from your decades old forum posts, newsgroups, old social media posts, silly newgrounds comments or helpful stackoverflow input, reddit shower thoughts, 9gag memes, etc.
Opting out now might prevent a negligible subset of your activity to be used in future training, but there will always be other services, other startups or flat out just companies who don't care about your consent.
The time to worry about your privacy online was probably 30 years ago, when the internet started picking up pace. More so around the 2010s when social media was booming.
All along, everyone knew you are handing your data, text and pictures to shady actors. Nobody cared.
Now these same actors give you a silly toggle and promise they won't use your data for AI training after virtually all leading AI models used this data to get to where we are. It's a clown show.
This is mostly futile regardless if it makes you feel better.
LLMs have been trained on everything from your decades old forum posts, newsgroups, old social media posts, silly newgrounds comments or helpful stackoverflow input, reddit shower thoughts, 9gag memes, etc.
Opting out now might prevent a negligible subset of your activity to be used in future training, but there will always be other services, other startups or flat out just companies who don't care about your consent.
The time to worry about your privacy online was probably 30 years ago, when the internet started picking up pace. More so around the 2010s when social media was booming.
All along, everyone knew you are handing your data, text and pictures to shady actors. Nobody cared.
Now these same actors give you a silly toggle and promise they won't use your data for AI training after virtually all leading AI models used this data to get to where we are. It's a clown show.
This assumes that apps and sites provider any toggles to turn this off.
Of course they don't. And those that do will definitely remove them at first opportunity.
The only hope is regulations like EU's DSA, but those are only as good as enforcement.
Of course they don't. And those that do will definitely remove them at first opportunity.
The only hope is regulations like EU's DSA, but those are only as good as enforcement.