Dodecahedron December

I try things on the internet.

rarely, shit just works.

  • 4 Posts
  • 143 Comments
Joined 1 year ago
cake
Cake day: July 5th, 2023

help-circle








  • Thanks for the reply. This sounds a lot like how AM / Talk radio works on the right as well. Leftist talk shows actually bring on interesting guests, but right wing shows always cut the mic when the guest says something thought-provoking.

    I would think conservatives would benefit from having their ideas challenged. It was the main reason why I started paying more attention to leftist ideas: they are actually defensible.






  • Gang, I hate to tell you this but this is what we mean when we say “you are the product” especially with free offerings.

    But if you hate that I have a worse thing to introduce you to: the internet. If you respond to this comment, or any comment on any lemmy instance or other federated service or website or blog… your words can be consumed, copied and used to train whatever anyone wants. It is trivially easy to create web scrapers with just a bit of coding knowledge. These days it’s pretty easy to then use that data to train AI models. To a computer, it’s just data.

    Grammarly is a product where you give it bad grammar and it gives you good grammar. Grammarly, like many products, gets better over time when it can understand what went wrong so its teams can make it right. This can often include any text entered into the program. I don’t know the specifics but they should be outlined in the privacy policy. A company using data it already has to train AI makes sense, especially if it anonymizes that data. It may not be ethical given that users weren’t aware of AI at the time they accepted the privacy policy, but with american capitalism a company can change a privacy policy and you can opt out if you don’t like it.

    That’s why we all have lawyers on retainer to read and translate all privacy policies for all websites and applications we interact with in a daily basis. Right? That’s normal, right?

    I will say, could this support person have meant that an organization with 500+ employees get a custom AI model trained on only the organization’s 500+ accounts? Because that would be better, and likely more ethical too.

    If that’s not the case and any content you have put into grammarly is being used to train AI, then I guess it’s time to stop using grammarly then huh? But it’s also time to stop posting anything on the web, too. Oh, and don’t publish anything, ever.

    Or, you could go with the flow. This data is mixed with millions of other accounts… sort of like what happened when chatgpt trained on anything you’ve already put out there. The only real concern I could see is if you discussed a very specific thing or invented your own personal coded style of writing and used it so much that, among the millions of other users, dominated the corpus and skewed the training model. Say there are only 5 grammarly users and you are number 5… you keep talking about “procorpia” being “mass sledge”, generating hundreds of entries with thousands of tokens “words”. By contrast let’s say the other 4 grammarly users only used it a few times a month to send short emails. Now, after training, the 6th grammarly user mispells a word as “procorpia” and grammarly generares “procorpia is totes mass sledge brah”. Suddenly, your secret is out.

    If, on the other hand you speak the same broken english as the rest of us, you are probably fine.







  • They were down but aren’t. This is going to happen from time to time for reasons, but most importantly (and this is not an advert or endorsement for centralized services like reddit):

    • these instances are run by small teams, maybe even one person per instance. By “run by” I mean the admins who can actually host and support the hosting environment of the instance, not moderators though that’s an important task too.
    • At reddit or other for-profit companies, multiple teams of people monitor multiple data centers worth of servers, have 24/7 tech support crew, dashboards, alarms, alerts, escallation proceedures drafted by other teams, people they can escallate problems to including usually a decent sized team at the physical datacenter due to the amount of servers they buy because of what they can afford based off advertising income because the site is popular enough, which is why it’s much more rare to see these services go down.

    But so many things can and do fail, including:

    • updates (dependencies, breaking updates, “this should just have worked but it didn’t, why?!”)
    • server issues (too many memes and now the disk has runeth over)
    • one server that gets overloaded or is in a data center that has a network failure, or a hardware failure on the server where the virtual server is hosted
    • account got hacked
    • 0 day exploit targeted directly at this server
    • DoS or DDoS attack
    • Admin has a day job that they need to do to keep the lights on at home and at the lemmy instance and has to do their day job work.

    Speaking from experience, but not with lemmy in particular.