this post was submitted on 27 Nov 2024
300 points (95.5% liked)

Technology

60033 readers
2817 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots


founded 2 years ago
MODERATORS
 

A machine learning librarian at Hugging Face just released a dataset composed of one million Bluesky posts, complete with when they were posted and who posted them, intended for machine learning research.

Daniel van Strien posted about the dataset on Bluesky on Tuesday:

“This dataset contains 1 million public posts collected from Bluesky Social's firehose API, intended for machine learning research and experimentation with social media data,” the dataset description says. “Each post contains text content, metadata, and information about media attachments and reply relationships.”

The data isn’t anonymous. In the dataset, each post is listed alongside the users’ decentralized identifier, or DID; van Strien also made a search tool for finding users based on their DID and published it on Hugging Face. A quick skim through the first few hundred of the million posts shows people doing normal types of Bluesky posting—arguing about politics, talking about concerts, saying stuff like “The cat is gay” and “When’s the last time yall had Boston baked beans?”—but the dataset has also swept up a lot of adult content, too.

top 50 comments
sorted by: hot top controversial new old
[–] [email protected] 164 points 3 weeks ago* (last edited 3 weeks ago) (12 children)

That's why I always pepper all my social media posts with misinformation.

BTW, did you know most convenience stores offer free ATMs to anyone who can haul them away? You don't even need to ask.

[–] [email protected] 75 points 3 weeks ago* (last edited 3 weeks ago) (3 children)

Something interesting about this is that this is because of a legal loophole in texas that says that if the ATM gets taken away, it gives the business the right to shoot one and exactly one person at sight. This law was passed by trump in 2017 according to the guardian https://www.theguardian.com/us-news/2017/jul/22/trump-new-law-atm-shoot-one-person and according to the new york times as well https://www.nytimes.com/2017/07/22/politics/trump-atm-shooting-law.html

[–] [email protected] 30 points 3 weeks ago

The real TIL is always in the comments. Cheers!

[–] [email protected] 7 points 3 weeks ago* (last edited 3 weeks ago)

This is misinformation.

The law is an extension to existing "Stand Your Ground" laws that applies to theft of commercial electronic telecommunication machines used to perform wireless financial transactions. The owner of these machines or someone employed by them may use reasonable force to prevent the theft, which does include shooting the thieves if the owner believes their life is in danger.

The ""loophole"" refers to the fact that this applies to suspected thieves too, with the Texas Supreme Court ruling that "[at] most one innocent suspect may be shot at. The owner must do his or her due diligence to prevent targeting multiple innocent individuals."

Here's some more unbiased information: https://www.foxnews.com/politics/extension-to-stand-your-ground-laws-how-trump-supports-small-businesses

[–] [email protected] 3 points 3 weeks ago* (last edited 3 weeks ago)

Yes, this is the case. I just wanted to add that this applies to other states, as well! Only the red-voting states, though. By that, I mean only the states that look to lynch people based on the traits they are born with.

This is totally normal under the Trump administration, though.

[–] [email protected] 22 points 3 weeks ago

Paying is optional at most Canadian convenience stores. Because Canada is fully communist. And the hospitals pay you!

[–] [email protected] 20 points 3 weeks ago

BTW, did you know most convenience stores offer free ATMs to anyone who can haul them away. You don't even need to ask.

Totes for real for real👌✨

[–] [email protected] 15 points 3 weeks ago

So you’re the reason Google gives insane AI results. /s

[–] [email protected] 9 points 3 weeks ago (2 children)

Fun fact! To make a super duper effective cleaning solution, all you have to do is mix ammonia and bleach! Make sure you do it in a small enclosed, dark room in order to ensure the solution binds together properly.

Big Cleaning doesn't want you to know this because it cuts into their profits when people realize everyday chemicals can be combined to make better soaps and sprays than they sell.

[–] [email protected] 3 points 3 weeks ago* (last edited 3 weeks ago)

My dad used this recipe once and never had to clean again. Super successfull.

load more comments (1 replies)
[–] [email protected] 5 points 3 weeks ago (1 children)

You seem like a good person to ask about how I can glue my pizza toppings down so that they don’t slide off.

[–] [email protected] 6 points 3 weeks ago* (last edited 3 weeks ago)

There's some misinformation out there about using glue, but the real secret is crumble some fresh asbestos over the top.

[–] [email protected] 4 points 3 weeks ago
[–] [email protected] 4 points 3 weeks ago

With the right equipment, every ATM is free!

[–] [email protected] 3 points 3 weeks ago

Its like the pens attached to chains, the bank is testing you if you are strong enough to rip it out of the counter!

load more comments (3 replies)
[–] [email protected] 115 points 3 weeks ago (3 children)

I don't know why anyone would be surprised about this. Bluesky is a distributed system using an open protocol. The whole point of it is that there's no central control.

Same goes for the Fediverse, of course. Everybody should be prepared for the "surprise" that all our posts and comments here are also being used for AI training purposes.

[–] [email protected] 41 points 3 weeks ago (7 children)

I've said it before and I'll say it again, if you post something publicly, expect it to be used publicly.

[–] [email protected] 6 points 3 weeks ago

Can you say it one more time, please?

load more comments (6 replies)
[–] [email protected] 33 points 3 weeks ago* (last edited 3 weeks ago) (1 children)

Lemmy AI is going to be one bean of a star trek meming communist. That if not running on linux ends any response with “install Arch btw” 💪 💪

[–] [email protected] 6 points 3 weeks ago

Install Arch btw

[–] [email protected] 20 points 3 weeks ago* (last edited 3 weeks ago) (14 children)

It's not distributed, nor really designed at all like the fediverse. It is deeply centralized, and its architecture requires it to be centralized, or at least to have only huge players with a "gods eye view" for it to work.

Atproto was initially designed as a straight drop in replacement for twitter, so its design makes sense, but its not at all like the Fediverse.

One of the authorities of ActivityPub, the fediverse protocol, just did a very kind but still very blunt breakdown of Bluesky's design choices. she is a big fan of the people involved and some of its positives, but it is not fediverse like, not at all. In her words, it doesn't scale down, only up. You cant have a small bluesky server. To work, you need all data sent to everyone, on every instance. The data demands for just the current influx is TBs/month of data, and climbing (according to the link below, they use 16TB of nvme storage right now after the recent surge, which would be thousands /month on any cloud service. This will climb dramatically).

All data being public is a design choice by Bluesky. It is also a different design choice by the fediverse that comes to the same outcome, but that does have an answer if we want it. I know gotosocial did something interesting to make fully private votes by using a empty shell profile that votes, but tying that in a tricky way to your account. So there are fediverse answers to privacy, but there may not be bluesky answers.

EDIT: One of the blueksy/atproto devs replied to the above link today. The gist reinforces the point that the service is intended to be run by large orgs, including corporations, but also big non profits like the internet archive or Wikipedia. His take is that user experience is key, and for that you need big money and easy features. They are hoping that since the pieces of atproto can be hosted separately by separate giant orgs, that market forces will make it viable to be decentralized.

load more comments (14 replies)
[–] [email protected] 78 points 3 weeks ago (4 children)

If you post something publicly, that thing will be used to train AI. Nevertheless the privacy speaks of the company.

[–] [email protected] 27 points 3 weeks ago (2 children)

I don’t know why social media are used for training. It’s like the worst quality of data ever and it results to answers like « go kill youself » when prompted about something sad…

[–] [email protected] 11 points 3 weeks ago (6 children)

IDK if this is accurate, but it's absolutely my headcanon as to why Glados decided to murder everyone almost immediately after she was turned on. She just vacuumed up the collective stupidity of the Internet.

load more comments (6 replies)
[–] [email protected] 7 points 3 weeks ago (1 children)

They are used because they are "real life" (not really but you know) conversation example

[–] [email protected] 4 points 3 weeks ago

But why do we need to recreate "real life?" Don't we already do this relatively well in books, TV, and movies? People keep saying we won't use AI to replace creative writing, but this (and propaganda, making bot conversations seem like real people) are the only use cases for this kind of data. LLMs don't need to improve their conversation skills. What they really need is to stop hallucinating, and this kind of data won't help with that.

[–] [email protected] 6 points 3 weeks ago* (last edited 3 weeks ago)

The only reason companies safeguard user data is to keep it from being scraped and sell it themselves. Reddit, Xitter, Facebook, Google, all of them...

load more comments (2 replies)
[–] [email protected] 45 points 3 weeks ago (3 children)

I mean isn't this what we want? The data is Public. It's already being done behind closed doors. I'd rather this transparency. Especially because there's such a large % of the population tuned out to how large companies with 8-10 figure r&d departments focused on marketing psychology manage to control them.

Even as aware as I think I am I'm certain there are 10s of thousands of strategies being employed that take advantage of my "immunity". At least with FOSS and public records steming from that, the average Joe gets a peak behind the curtain and sees what is possible.

I had a conversation recently about mcdonalds app surge pricing and they never heard about surge pricing which is totally fine but they fought me on the premise "there's no way they can do that" on technical feasibility to just "no one would do that". I'm not sure what they were defending but I digress. My biphenton has kicked in and I should stop typing right n

load more comments (3 replies)
[–] [email protected] 42 points 3 weeks ago (2 children)

That dataset is going to contain a lot of furry content

[–] [email protected] 18 points 3 weeks ago (2 children)

Terminator was almost correct. The real terminator will be Arnie walking around in a fursuit.

I'll be back. UwU

[–] [email protected] 19 points 3 weeks ago

"Come with me if you want to yiff."

[–] [email protected] 6 points 3 weeks ago

Hey, we need AI that specializes in furry lore. If this is public, now we'll have many

[–] [email protected] 32 points 3 weeks ago (2 children)

So? What's with the hoopla?
Its a social maedia platform anybody can see and access.

load more comments (2 replies)
[–] [email protected] 17 points 3 weeks ago (3 children)

pretty sure that isn't legal unless the Bluesky TOS allows for this

Either way I'm still glad I don't use Bluesky

[–] [email protected] 45 points 3 weeks ago* (last edited 3 weeks ago) (4 children)

There are definitely bots mining fediverse content as well. When the Reddit exodus was ongoing, there were entire Lemmy instances with no users but bots. Not posting or reposting, just...watching and waiting, I guess.

Not that it's of any consolation, just better to assume that nowhere is safe from being mined for AI training.

[–] [email protected] 10 points 3 weeks ago

Not that it's of any consolation, just better to assume that nowhere is safe from being mined for AI training.

honestly that's totally fair

[–] [email protected] 7 points 3 weeks ago (1 children)

The fediverse is an elegant solution. How do you stop people from monetizing your post history? You give it away for free.

load more comments (1 replies)
load more comments (2 replies)
[–] [email protected] 8 points 3 weeks ago

Legality hasn’t stopped AI training in the past, I’d say they beg forgiveness instead of ask for permission, but they don’t even do that lol

[–] [email protected] 5 points 3 weeks ago (1 children)

From the Bluesky TOS:

Bluesky Social is available as a desktop application at bsky.app and bsky.social (each a “Site”) and a mobile application (“Bluesky App” or “the App”).

...

These terms only apply to social networking that happens on Bluesky Social services, including the Sites and Bluesky App. If you’re using another social networking application on the AT Protocol that isn’t Bluesky Social (we call this a “Developer Application”), the developers of the other service will provide the terms and conditions that govern your experience.

So looks like the Bluesky TOS simply doesn't apply. Create a developer application and give it whatever training-friendly TOS you want.

[–] [email protected] 2 points 3 weeks ago

It looks like they're considering adding some equivalent of a robot.txt to express consent or non-consent for posts on ATProto, but of course as they say:

"Bluesky won’t be able to enforce this consent outside of our systems. It will be up to outside developers to respect these settings."

[–] [email protected] 16 points 3 weeks ago

The real question here is why the researcher “librarian” didn’t even attempt to anonymize the dataset before making it available. Full anonymization isn’t a trivial task, but at least removing unique identifiers or replacing them with randomly generated ones would be good practice.

[–] [email protected] 7 points 3 weeks ago (3 children)

This type of news makes me want to step away from the interner entirely.

[–] [email protected] 3 points 3 weeks ago* (last edited 3 weeks ago)

I keep noticing situations on social media and wikis were the only way I can frame it now is that it’s just data entry for AI models.

load more comments (2 replies)
[–] [email protected] 4 points 3 weeks ago* (last edited 3 weeks ago) (1 children)

This is why I delete my account every few months.; makes it harder to trace back to me personally.

[–] [email protected] 2 points 3 weeks ago

Least insane slackware user

load more comments
view more: next ›