Live data from Hacker News

Can I remove my personal data from GenAI training datasets?

knowingmachines.org

51–60 of 123 posts

Re: Can I remove my personal data from GenAI training datasets?

#51

I don't get it, you put that information on the internet, you have no expectation to privacy. But now maybe you learned a lesson, and you won't publicly share things you don't want people to see?

> I don't get it, you put that information on the internet, you have no expectation to privacy.

While that is somewhat true, there are multiple angles to it. The big one is expectations. It used to be that if you shared something on an obscure forum that had like 100 readers, most of those people expected that only 100 would see their posting. And it was true, so it reinforced the expectation for nontechnical people, which is basically everyone within a rounding error.

Then everything gets indexes and now fed to AI and suddenly billions of people have easy access to what they felt was very limited in distribution.

You could file this under "People don't threat model correctly", but also (more importantly) under the fact that technology should consider human nature before breaking expectations.

Re: Can I remove my personal data from GenAI training datasets?

#52

I don't get it, you put that information on the internet, you have no expectation to privacy. But now maybe you learned a lesson, and you won't publicly share things you don't want people to see?

> you put that information on the internet

That isn't necessarily true. Someone else might have put the information on the Internet. That could be someone else uploading a photo of you, or records of your home purchase on a city's public records, or an obituary or wedding announcement that listed you as family, etc.etc.

Re: Can I remove my personal data from GenAI training datasets?

#53

I don't get it, you put that information on the internet, you have no expectation to privacy. But now maybe you learned a lesson, and you won't publicly share things you don't want people to see?

There are a lot of aspects to this though. We write code for solar power plants, some of it is open source or at least publicly available. What if we wrote something that turned out to be bad and it was used by someone else through the AI and it broke their plant… would we be responsible?

Now, you’re probably thinking about this from a reasoning or technical perspective in which case it’ll appear to be a ridiculous concern… because it is a ridiculous concern. That’s not how our legal department sees it though. They see it as risk mitigation, and they actually take it rather serious.

Re: Can I remove my personal data from GenAI training datasets?

#54
post #23

Earlier quoted context omitted.

For most companies “deleting the model” is equivalent to dissolving the company so that is equivalent to not being able to comply. More realistically what they would need to do is exit the market of the country that has such stupid laws.

>For most companies “deleting the model” is equivalent to dissolving the company I'm okay with that. It's sad that you're not. If you're willing to start a business on such shady foundations, there's a really good chance your business will continue to make shady decisions in the future. It's better to find and remove the cancer early

I just disagree that it's shady. If you put your personal info into public circulation I think it's absolutely morally fine to have a model train on it.

Re: Can I remove my personal data from GenAI training datasets?

#55

Earlier quoted context omitted.

> you put that information on the internet, you have no expectation to privacy That was the expectation until Zuckerberg came along. People were putting tons of stuff on BBSs, intranets, and other stuff 20+ years beforehand because people had the idea of "consent" back then existed: using someones information for a purpose they didn't intend was (and still is) wrong.

Using someone's information for a purpose they didn't intend is wrong? What? This just seems obviously mistaken. I don't agree with that at all. I'm not even sure how to argue about that. Lots of art, science and technology can be considered as "using someone's information for a purpose they didn't intend." It is extremely normal for people to find new uses for things; information is not exempt from this.

Consider the thought experiment that you give your postal address to some business, because you want to subscribe to regular grocery deliveries. Then you notice that each delivered package contains a small transparent bag with some with powder in it.

Whoever treats others' information can only do so with a clear purpose and for a defined time period according to current EU laws.

If, however, you were referring to information that you published on the Internet for everyone's benefit then you would still need to consider intellectual property rights. In the open source software world we have the licenses that deal with this, and then there are copyright laws protecting content providers (not making a case here whether they are good or not).

I guess what makes a difference is if there is some business involved either in the production or in the consumption side of the equation and if we accept that "machine learning" is the same as "human learning".

EDIT: separated paragraphs, typo

Re: Can I remove my personal data from GenAI training datasets?

#58
post #51

I don't get it, you put that information on the internet, you have no expectation to privacy. But now maybe you learned a lesson, and you won't publicly share things you don't want people to see?

> I don't get it, you put that information on the internet, you have no expectation to privacy. While that is somewhat true, there are multiple angles to it. The big one is expectations. It used to be that if you shared something on an obscure forum that had like 100 readers, most of those people expected that only 100 would see their posting. And it was true, so it reinforced the expectation for nontechnical people,…

You should consider the size of the training set as well. A blog post in a 30T dataset like RedPajama has less impact than one in a fine-tuning dataset of 1000 examples. The gradients from all tokens are added up, they stack on top of each other and their influence is diluted in larger datasets.

Re: Can I remove my personal data from GenAI training datasets?

#59
post #55

Earlier quoted context omitted.

Using someone's information for a purpose they didn't intend is wrong? What? This just seems obviously mistaken. I don't agree with that at all. I'm not even sure how to argue about that. Lots of art, science and technology can be considered as "using someone's information for a purpose they didn't intend." It is extremely normal for people to find new uses for things; information is not exempt from this.

Consider the thought experiment that you give your postal address to some business, because you want to subscribe to regular grocery deliveries. Then you notice that each delivered package contains a small transparent bag with some with powder in it. Whoever treats others' information can only do so with a clear purpose and for a defined time period according to current EU laws. If, however, you were referring to inf…

Copyrights protect the copying of the original text, but models take gradients. Are gradients protected as well? Even when data is copyrighted it still has legitimate value for training, pure ideas don't get copyright protection, only expression.

Re: Can I remove my personal data from GenAI training datasets?

#60

I don't get it, you put that information on the internet, you have no expectation to privacy. But now maybe you learned a lesson, and you won't publicly share things you don't want people to see?

> you put that information on the internet, you have no expectation to privacy That was the expectation until Zuckerberg came along. People were putting tons of stuff on BBSs, intranets, and other stuff 20+ years beforehand because people had the idea of "consent" back then existed: using someones information for a purpose they didn't intend was (and still is) wrong.

> That was the expectation until Zuckerberg came along.

The early days of the Internet were full of "information should be free" hackers, many of whom roam this very forum in their 40s, 50s, and 60s.

Well, information is free now. And this is what it looks like - consumed by LLMs. We got what we asked for.

Post reply on HN