Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Updates to Consumer Terms and Privacy Policy

anthropic.com

Post reply on HN
Print view View original

521–530 of 551 posts

Re: Updates to Consumer Terms and Privacy Policy

#521

Post by ljosifov »

12ian34 wrote: post #356
ljosifov wrote: post #350

Excellent. What were they waiting for up to now?? I thought they already trained on my data. I assume they train, even hope that they train, even when they say they don't. People that want to be data privacy maximalists - fine, don't use their data. But there are people out there (myself) that are on the opposite end of the spectrum, and we are mostly ignored by the companies. Companies just assume people only ever w…

not remotely worried about leaks, hacks, or sinister usage of your data?

I'm worried, it's not like I don't care. For example, I'm worried that Google is such a huge ginormous target, that at some point their Gmail will be broken. At the same time, there are benefits to sharing data. There are benefits to me, in Google using the information it has on my, to make my life easier. In this case, I judge that Gemini using my data to train, is a low extra risk for me. Compared to all other risks I take, for doing things in public. Including writing this on public forums, as you do too.

In general, I find the ongoing public scare about sharing data, to be anti-thesis to the original spirit of the Net, that was all about sharing data. Originally, we were delighted to connect to perfect strangers on the other side of the world. That we would never have gotten to communicate with otherwise. I accept there might have been an element of self-selection there, that aided that view: people one'd communicate with, although maybe from a different culture, would be from similar niche sub-culture of people messing with computers and looking forward to communication, having a favourable view of that.

Re: Updates to Consumer Terms and Privacy Policy

#522

Post by ljosifov »

int_19h wrote: post #486
ljosifov wrote: post #400

Earlier quoted context omitted.

If they leaked bank accounts numbers, or private keys - I would be worried. That has not happened in the past. About myself personally - my Name Surname is googleable, I'm on the open electoral register, so my address is not a secret, my company information is also open in the companies register, I have a a personal website I have put up willingly and share information about myself there. Training models on my data d…

LLMs can and do sometimes regurgitate parts of training data verbatim - this has been demonstrated many times on things ranging from Wikipedia articles to code snippets. Yes, it is not particularly likely for that damning private email of yours to be memorized, but if you throw a dataset with millions of private emails onto a model, it will almost certainly memorize some of them, and nobody knows what exact sequence…

That's a consideration, for sure. But given the LLM-s have not got the ground truth, everything is controlled hallucination, then - if the LLM tells you an imperfect version of my email or chat, you can never be sure if what the LLM told you is true, or not. So maybe you don't gain that much extra knowledge about me. For example, you can reasonably guess I'm typing this on the computer, and having coffee too. So if you ask the LLM "tell me a trivial story", and LLM comes back with "one morning, LJ was typing HN replies on the computer while having his morning coffee" - did you learn that much new about me, that you didn't know or could guess before?

Re: Updates to Consumer Terms and Privacy Policy

#523

Post by ljosifov »

j4hdufd8 wrote: post #365
ljosifov wrote: post #350

Excellent. What were they waiting for up to now?? I thought they already trained on my data. I assume they train, even hope that they train, even when they say they don't. People that want to be data privacy maximalists - fine, don't use their data. But there are people out there (myself) that are on the opposite end of the spectrum, and we are mostly ignored by the companies. Companies just assume people only ever w…

> But there are people out there (myself) that are on the opposite end of the spectrum, and we are mostly ignored by the companies. Companies just assume people only ever want to deny them their data. What? I think you're exactly the kind of person that companies pay attention to, and why they pull moves like this

My experience in the UK medical systems has been the opposite - wrote here

https://news.ycombinator.com/item?id=45066321

Google knows what "Home" is for me only in Gmaps, because I went out of my way (put a Label etc) to tell it. I want to be able to tell Google "My home is XYZ", and for Google to use that information about me in all of Google ecosystem. When I talk to Gemini it should know what/where "LJ home" is, when I write in Gdoc it should know my home address (so to insert it if I want it), ditto for Gmail, when I search in Google photos "photos taken at home" it should also know what "home" is for me.

I have the impression that we ended up in the worst case scenario. People I don't want to have my data, have access to it. People I do want to have my data, are afraid to touch it, and use it - yes! - for theirs, but also for my benefit too.

Re: Updates to Consumer Terms and Privacy Policy

#524

Post by ljosifov »

bgwalter wrote: post #401
ljosifov wrote: post #350

Excellent. What were they waiting for up to now?? I thought they already trained on my data. I assume they train, even hope that they train, even when they say they don't. People that want to be data privacy maximalists - fine, don't use their data. But there are people out there (myself) that are on the opposite end of the spectrum, and we are mostly ignored by the companies. Companies just assume people only ever w…

I realize this might be satire. If not, you are using the same aggressive strategy of turning the tables as Palantir: https://www.theguardian.com/technology/2025/jul/08/palantir-... Most people do want to deny their data, as we have recently seen in various DOGE backlashes.

It's not a satire, you can check mu comments on this topic easily.

I dispute 'most people'. Revealed preferences of most people are that they value their data privacy very cheaply, almost zero. Even one click extra to share their data less, is one click too many, an effort too high - for most people. This is their real, observed behaviour. I think our current predicament is the case of "public lies, private truths." A small cadre of vocal proponents of a particular view, established "the ground truth to what is desirable". (in this case - maximum privacy, ideally zero information sharing) The public goes with it in words, pays lip service - but in reality behaves different, even opposite to what they say they desire.

And even if 'most people' wanted what you say they do, I still think the companies could and should accommodate a minority group like myself that want otherwise to what 'most people' want. I don't think the will of the majority is the highest ideal, so high as to trump what I personally want.

Re: Updates to Consumer Terms and Privacy Policy

#525

Post by ljosifov »

calmbonsai wrote: post #391
ljosifov wrote: post #350

Excellent. What were they waiting for up to now?? I thought they already trained on my data. I assume they train, even hope that they train, even when they say they don't. People that want to be data privacy maximalists - fine, don't use their data. But there are people out there (myself) that are on the opposite end of the spectrum, and we are mostly ignored by the companies. Companies just assume people only ever w…

I don't think you understand how...humanity works?! Is this deliberate parody? Abuse of medical data is just the tip of the iceberg here and, at least in the states, privatized healthcare presents all sorts of for-profit pricing abuse scenarios let alone nasty scenarios for social coercion.

You know little about me, so it's better to assume less, no? My personal experience with medical data specifically is, that I would have been harmed by obstacles to data sharing that the UK medical system has in place, having not been familiar with computers and tech enough to anticipate the ways lack of data sharing will lead to outcomes undesirable to me. I wrote about that in a comment here https://news.ycombinator.com/item?id=45067219

Re: Updates to Consumer Terms and Privacy Policy

#526

Post by ljosifov »

danparsonson wrote: post #509
ljosifov wrote: post #488

Earlier quoted context omitted.

Whether Google is interested in serving me or not, is not only untestable (i.e. what counts as 'Google', 'interested', and 'serving' there - one could argue to end of time) - but besides the point. I want to be able to tell Google "My home is XYZ", and for Google to use that information about me in all of Google ecosystem. When I talk to Gemini it should know what/where "LJ home" is, when I write in Gdoc it should kn…

You started by saying that it's difficult or impossible to define what 'serving the user' looks like, then immediately gave examples of what it would look like to you. It's not that Google can't do these things or is afraid to, but rather that operating in your best interests does not benefit their shareholders. Sure, it'd be great if we could all just get along, but we're living in the worst case scenario you descri…

Yes - I meant 'impossible to difficult' to define to all people, at all times. Agree it's easy for me to define how that looks. It doesn't mean that the same is true to you. That's why I went from a very general, to very specific.

I'm saying we ended up in situation where people are lying when they say "I don't trust Google", b/c they have Gmail, use Google services - so their trust can't be zero. It's more than zero. Obviously it's a trade-off, people are pragmatic they do their cost-benefit analysis, and act accordingly. They just lie when they talk about the subject. I think it'd be better for all, if the public discussion moved from "I trust Google zero" (which is obviously untrue), to "There is cost-benefit to this, and I personally chose xyz".

Re: Updates to Consumer Terms and Privacy Policy

#527

Post by 12ian34 »

ljosifov wrote: post #521
12ian34 wrote: post #356

Earlier quoted context omitted.

not remotely worried about leaks, hacks, or sinister usage of your data?

I'm worried, it's not like I don't care. For example, I'm worried that Google is such a huge ginormous target, that at some point their Gmail will be broken. At the same time, there are benefits to sharing data. There are benefits to me, in Google using the information it has on my, to make my life easier. In this case, I judge that Gemini using my data to train, is a low extra risk for me. Compared to all other risk…

> the ongoing public scare about sharing data

I think this might be a bit of a social bubble thing - I think it isn't a forefront concern for the vast majority of people.

Re: Updates to Consumer Terms and Privacy Policy

#528

Post by ljosifov »

12ian34 wrote: post #527
ljosifov wrote: post #521

Earlier quoted context omitted.

I'm worried, it's not like I don't care. For example, I'm worried that Google is such a huge ginormous target, that at some point their Gmail will be broken. At the same time, there are benefits to sharing data. There are benefits to me, in Google using the information it has on my, to make my life easier. In this case, I judge that Gemini using my data to train, is a low extra risk for me. Compared to all other risk…

> the ongoing public scare about sharing data I think this might be a bit of a social bubble thing - I think it isn't a forefront concern for the vast majority of people.

I think you are correct there - the majority of the public don't care. They just try to get about doing their daily business and act the best they can under circumstances. So we just click "Accept" to any popup banner make it go away, accept "All cookies" 100 times every day, use Google mail/map/photos/drive and that all involves giving away data, even if in words we say we don't want to give data. So yes obviously the public by necessity act in a rational way, doing cost-benefit analysis. While a cadre of privacy obsessives have made my life worse by lobbying and having their bad ideas codified in the UL laws. Wrote about my experience in the UK medical systems here https://news.ycombinator.com/item?id=45066321

Re: Updates to Consumer Terms and Privacy Policy

#529

Post by bluecalm »

Deegy wrote: post #338
bluecalm wrote: post #328

Earlier quoted context omitted.

They might optimize learning to weight novel/unexpected parts more in the future. The better the models become (the more the expect) the more value they will get from unexpected/new ideas.

Good point. But can the models even behave that way? They depend on probability. If they put a greater weight on novel/unexpected outputs don't they just become undependable hallucination machines? Despite what some people think, these models can't reason about a concept to determine it's validity. They depend on recurring data in training to determine what might be true. That said, it would be interesting to see a m…

I think it's happening already. Chat GPT was able to connect my name to my project based on chess.com profile and one Hacker News post for example. It's not that hard to imagine that it learns a solution to a rare problem based on one input point. It may see one solution 1000 times an a rare solution 1 time and it can still be able to reference both.

Re: Updates to Consumer Terms and Privacy Policy

#530

Post by handoflixue »

elashri wrote: post #77
JohnnyMarcone wrote: post #25

Earlier quoted context omitted.

I got a pop-up when I opened the app explaining the change and an option to opt out. That seems very transparent to me.

> That seems very transparent to me Implicit consent is not transparent and should be illegal in all situations. I can't tell you that unless you opt out, You have agreed to let me rent you apartment. You can say analogy is not straightforward comparable but the overall idea is the same. If we enter a contract for me to fix your broken windows, I cannot extend it to do anything else in the house I see fit with Implic…

How is it "implicit" to click "I agree" to a large pop-up that takes up most of the screen?
Post reply on HN