why would I want to use it?

You wouldn’t, but that’s fine with Match Group: JP Morgan^[1] are loving this new monetization strategy. If they think they can get more money out of their users they will, the experience and usefulness of their app be damned. Very similar to aggressively monetized mobile games, but extra icky since they’re monetizing human relationships.

Agreed. Strong (and effectively enforced) worker protections are just as important as tech-specific safety regulations. Nobody should feel like they need to put themselves into a risky situation to make work happen faster, regardless of whether their employer explicitly asks them to take that risk or (more likely) uses other means like unrealistic quotas to pressure them indirectly.

There are certainly ways to make working around robots safer, e.g. soft robots, machine vision to avoid unexpected obstacles in the path of travel, inherently limiting the force a robot can exert, etc… And I’m all for moving in the direction of better inherent safety, but we also need to make sure that safer systems don’t become an excuse for employers to expose their workers to more risky situations (i.e. the paradox of safety).

This is really cool, thanks for sharing!

It seems like you’re working under the core assumption that the trained model itself, rather than just the products thereof, cannot be infringing?

Generally if someone else wants to do something with your copyrighted work – for example your newspaper article – they need a license to do so. This isn’t only the case for direct distribution, it includes things like the creation of electronic copies (which must have been made during training), adaptations, and derivative works. NYT did not grant OpenAI a license to adapt their articles into a training dataset for their models. To use a copyrighted work without a license, you need to be using it under fair use. That’s why it’s relevant: is it fair use to make electronic copies of a copyrighted work and adapt them into a training dataset for a LLM?

You also seem to be assuming that a generative AI model training on a dataset is legally the same as a human learning from those same works. If that’s the case then the answer to my question in the last paragraph is definitely, “yes,” since a human reading the newspaper and learning from it is something that, as you say, “any intelligent rational human being” would agree is fine. However, as far as I know there’s not been any kind of ruling to support the idea that those things are legally equivalent at this point.

Now, if you’d like to start citing code or case law go ahead, I’m happy to be wrong. Who knows, this is the internet, maybe you’re actually a lawyer specializing in copyright law and you’ll point out some fundamental detail of one of these laws that makes my whole comment seem silly (and if so I’d honestly love to read it). I’m not trying to claim that NYT is definitely going to win or anything. My argument is just that this is not especially cut-and-dried, at least from the perspective of a non-expert.

Well I hear what you’re saying, although I don’t much appreciate being told what I should want the outcome to be.

My own wants notwithstanding, I know copyright law is notoriously thorny – fair use doubly so – and I’m no lawyer. I’d be a little bit surprised if NYT decides to raise this suit without consulting their own lawyers though, so it stands to reason that if they do indeed decide to sue then there are at least some copyright lawyers who think it’ll have a chance. As I said, we’ll see.

Yeah I’ve heard a lot of people talking about the copyright stuff with respect to image generation AIs, but as far as I can see there’s no fundamental reason that text generating AIs wouldn’t be subject to the same laws. We’ll see how the lawsuit goes though I suppose.

Installed! I’ve been unhappy with my weather app for a little while now, looking forward to giving this one a try! The fact that they’ll use an approximate location is really nice. Thank you for shouting this out!

You can send downvotes using 3rd party clients, but beehaw doesn’t register or track them. Hitting the button does nothing (and it’s not even present in the web ui)

Really interesting writeup, thank you for sharing! Many of the technical details go well over my head but nonetheless it’s very interesting to hear some of these success stories, and it also sheds light on how much work running an instance with a lot of users actually is. Here’s hoping that future versions of lemmy with (eg) more optimized database code will make life easier for all the folks in the operations team!

Yeah that makes sense, I’m not a mod in this community. If you go to one of my comments on /c/science it should display I think (although it’s been a bit since I tried using Jerboa)

Their actual username is xtremeownage, which is why it still links that way, but lemmy lets users set a display name that’s different from their username (in this case, HTTP_404_NotFound). Sounds like kbin doesn’t respect lemmy display name settings.

I’ll hop on just to add: the little plate that says “mod” can’t be turned off, but for us community moderators we can toggle a little shield icon when we’re trying to speak with the “mod voice.” We can only do it in our communities, but (I think) the admin team is by default modded in every community and should be able to toggle that shield wherever they are.

So if you see the shield icon next to the name, that was intentionally toggled on to show they’re trying to speak as a mod and not just a community member. If you only see the little “mod” nameplate you can probably assume they’re just talking as a member of the community (unless context makes it obvious that they just forgot to turn on the shield).

Edit: Here’s an example of how the shield icon looks to me from one of my posts; I’m on mobile with the darkly theme so it probably looks different to you unless you’re also on mobile, but hopefully still a useful example:

I’ve not used requests, but yes their docs make it look like it really is that easy: https://requests.readthedocs.io/en/latest/user/quickstart/

Looks like the .json() call just returns a dictionary (or maybe a list of dictionaries), which means you can use all of python’s normal dictionary methods to find the data you’re looking for!

Yeah as far as I can tell without a close read of the paper – which I just don’t have time to do – it’s a review paper that just happens to have been written by some of the people involved in this commercial endeavor. (I imagine they don’t want to share their proprietary composition and architecture.) It seems like the reporter covering this didn’t catch the distinction.

To my previous comment, this means we really have no idea if they’re using a lead-containing composition or not. Like I mentioned, perovskites are a large class of materials. Since the review paper has a section on the challenges around lead containing compositions I’d hope they’re not using lead, but who knows 🤷‍♂️

I feel like I should clarify because the article didn’t do a good job at explaining: perovskite is a kind of structure, not a particular material. They have the generic formula ABX₃ (where A and B are different kinds of cations and X is some kind of anion), although not everything with the formula is a perovskite.

Simple perovskites include some lead-containing materials like lead titanate, but also lead-free materials like barium or strontium titanate. And in general there are a lot of different kinds of perovskites, especially because some of the structural sites can be filled by small organic molecules instead of pure elements.

Edit: I think I was misreading the journal article before my edit (it’s early I’m not awake yet lol). I had said it looks like they’re using a lead-based perovskite but actually I can’t tell what exactly they’re using with a quick skim. The article is very review-y, the formula I thought they were using is from another paper. I’ll have a more thorough look later.

Edit 2: It’s a review paper, and the way it was described in the linked news article is kinda misleading. Its not specifically about this company’s particular composition or architecture.