AnAmericanPotato

joined 9 months ago
[–] AnAmericanPotato@programming.dev 21 points 1 week ago (4 children)

I’m not (currently) in a position where others would find it desirable to do so. Potentially in the future?

It's hard to imagine a scenario where this would happen and your voice would not otherwise be available. For example, if you went into politics, then you'd be a target, but you'd already be speaking in public all the time. It only takes a few seconds of a voice sample to do this nowadays and it'll only get easier from here.

Maybe just make a point to educate your family and friends on the risk of voice cloning so they don't fall for phone scams.

I've noticed an uptick as well. This isn't the first time it's happened over the years, though. Spam is a cat-and-mouse game. Every now and then spammers learn how to break through, and it takes some time for Google to adapt.

I've been surprised by the latest wave, because it's so obviously spam. Mostly phishing attempts full of misspellings and even numbers in place of letters, like F1del1ty instead of Fidelity. Should be pretty easy to filter.

[–] AnAmericanPotato@programming.dev 21 points 1 week ago (7 children)

The article keeps referring to "the sport", but never mentions any particular sport. I don't think I've ever seen "the sport" used to refer to sports in general before. Is this a regional language difference, like how Americans would say "go to the hospital" while Brits would say "go to hospital"?

"Let"

As if anyone asked for this.

I think you're overestimating how many people keep their eyes open at all.

[–] AnAmericanPotato@programming.dev 65 points 2 weeks ago* (last edited 2 weeks ago) (6 children)

Racism in America is real. Anyone telling you otherwise is probably just living a charmed life and incapable of accepting that their personal experience is not universal.

I don't have the time or energy to prove this exhaustively, but here's a starting point: https://en.wikipedia.org/wiki/Driving_while_black

In 2019, as reported by NBC, the Stanford Open Policing Project found that "police stopped and searched black and Latino drivers on the basis of less evidence than used in stopping white drivers, who are searched less often but are more likely to be found with illegal items."

Please refer to the citations on that page for more details. Lots of studies in various states showing the same thing. The fact that the mere existence of racial profiling in America is still debated, when it has been consistently proven again and again for decades, is itself a clear indicator of a different kind of racism.

Here's a little story that stuck in my memory, about how a white woman finally came to realize that racial harassment by police was a real thing. It's kind of hilarious, in a dark, face-palmy kind of way. https://franklywrite.com/2020/06/01/a-white-woman-racism-and-a-poodle/

It’s explicitly forbidden for anyone to discriminate against you based on your race or ethnicity

Ironically, it's very common to be asked for this information specifically because of anti-discrimination laws, so they can demonstrate statistically fair practices. I always see a box for this on medical forms, new-hire paperwork, etc. I believe the law requires it to be optional and only used for regulatory reports. So that's probably what OP heard about.

Oh nice! I did a quick search for forks but didn't find that, so thanks for linking. I'll check it out!

[–] AnAmericanPotato@programming.dev 9 points 2 weeks ago (1 children)

This assumes a legitimate need to prove who you are outside the context of that specific site, rather than just within it. Sometimes that need is real, sometimes it is not.

When it's not, and you only need to prove you are the same person who created the account, then a simple username and password is sufficient. Use 2FA (via authenticator app or key, NOT via SMS or email) on top of that. This allows users to prove to a sufficient degree that they are the owner of that account.

This is how most Lemmy instances work, for example. I can sign up by creating a username and password, with optional 2FA. They do not need my email. They do not need my phone number. They do not need my name, or my contacts, or anything else that is not related to my identity within their server.

I realize that this is untenable at large scales for any communications platform. Spam (and worse) is a problem wherever there are easy and anonymous signups. I'm honestly not sure how Lemmy is as clean as it is. I guess it's just not popular enough to attract spammers.

[–] AnAmericanPotato@programming.dev 1 points 2 weeks ago (2 children)

Looks like the open source version was axed, and the cheapest subscription plan is $36.67 per month per user. Yikes.

[–] AnAmericanPotato@programming.dev 2 points 2 weeks ago (4 children)

I'd kill for a modern HyperCard.

which would indicate that it’s somehow needed to generate AI-generated CSAM

This is not strictly true in general. Generative AI is able to produce output that is not in the training data, by learning a broad range of concepts and applying them in novel ways. I can generate an image of a rollerskating astronaut even if there are no rollerskating astronauts in the training data.

It is true that some training sets include CSAM, at least in the past. Back in 2023, researches found a few thousand such images in the LAION-5B dataset (roughly one per million images). 404 Media has an excellent article with details: https://www.404media.co/laion-datasets-removed-stanford-csam-child-abuse/

On learning of this, LAION took down their database until it could properly cleaned. Source: https://laion.ai/notes/laion-maintenance/

Those images were collected from the public web. LAION took steps to avoid linking to illicit content (details in the link above), but clearly it's an imperfect system. God only knows what closed companies (OpenAI, Google, etc.) are doing. With open data sets, at least any interested parties can review, verify, and report this stuff. With closed data sets, who knows?

view more: next ›