In the DOGE case, they specifically broke all the controls that existed to manage insider risk and keep people from making copies like this, but (especially 20-30 years ago) I've been on plenty of networks that just had no concept of insider risk and everything was just open for anyone to access (or protected by shared passwords everyone knew).
So you're saying that if you worked there you would also steal the social security data? What am I supposed to be taking away from this besides the fact that you would make poor choices and lack ethics? Didn't seem like it was a problem for people who worked in gov't prior to DOGE existing, so I'm not really getting any other takeaway here.
But look around the network, see what file shares are world readable, maybe see if there's any FTPs or Telnet servers with no username/password (or at least, no password stronger than "guest")? That's just being curious. And if I see any interesting files, and I make a copy to look at later, that's not a crime, is it?
I'd like to think my younger self, if he'd been hired at the SSA or somewhere similar, would see the difference between "the personal data of hundreds of millions of people" and the networks I actually had access to at the time. I know I wouldn't be trying to sell the data or trying to otherwise leverage it for financial gain, but I don't really have such a high opinion of my younger self's judgement that I would completely rule out making a copy for objectively dumb reasons.
I don't know about this person specifically, but the news from when DOGE was active was full of "employee of department fired for trying to prevent DOGE employees from directly accessing system no one is allowed to directly access".
Crazy that of all places someone here will say that the volume has nothing to do with the platform? The entire point of the platform is its size. The entire point of the tech industry is its size...
what would you specifically like to happen, and how?
my supposition is about an efficient use of energy. people are spending a lot of energy inefficiently by trying to associate this issue - legally - with Roblox. I'm saying that people are asking the wrong questions and getting answers to those wrong questions, all while the problem persists because its happening elsewhere while being conflated with Roblox. All while despite the legal limitations, Roblox already has implemented changes far above and beyond what's required, while the problem still doesn't go away for a very specific reason of it not occuring on Roblox.
They need to hire a sufficient amount of human monitors, I don't think it's that complicated at all. And people have to spend that energy because Roblox keeps connecting children with pedophiles.
> All while despite the legal limitations, Roblox already has implemented changes far above and beyond what's required, while the problem still doesn't go away for a very specific reason of it not occuring on Roblox.
I keep seeing people write this almost verbatim all over this thread.
I haven't read the rest of the thread, I respond from my comment history, so if people are writing the same thing, maybe there's some truth to it, are you familiar with the user experience on Roblox?
where many people can't chat with each other at all, but they can click to an external link or view instructions in an experience to join discord?
in this case, how would you modify the user experience?
I'd like to hear your reasons for this. From what I see, most discussions on child exploitation on Roblox eventually make clear that the exploitation happened outside of Roblox, most commonly on Discord. Moreover, it's a lot more difficult to censor a few messages talking about alternative platforms than it is to stop a long-running logged chat conversation where child exploitation usually takes place.
Roblox connects them in the first place, what does it matter that it progresses outside of Roblox? That would necessarily be the case if they were to ever meet IRL anyway.
Oh it's difficult for Roblox? A $42billion company? Whose entire business model is based around kids? It's difficult for them?? Boo fucking hoo.
The problem is that the external platforms that the children progress to generally have much laxer protection systems than Roblox does, and thus end up more vulnerable. I just chose Discord as an example as they're the most commonly cited chat platform that exploitation beginning on Roblox ends up on, and they also have problems with their trust & safety team that allows this to occur.
Meeting IRL is a problem as well, it just makes up fewer of the cases.
>The problem is that the external platforms that the children progress to generally have much laxer protection systems than Roblox does, and thus end up more vulnerable.
So? Discord is a problem too. But they aren't finding the kids on discord because Discord is not a social network that links pedophiles with children. Roblox is that.
>Meeting IRL is a problem as well, it just makes up fewer of the cases.
Again... and? By your logic IRL is the problem too because for some reason you think we should not expect Roblox to do anything about the fact that it connects children with pedophiles. But IRL isn't a platform. And if Roblox was IRL, it would've already been sued into oblivion because it facilitates pedophiles predating on children.
> But they aren't finding the kids on discord because Discord is not a social network that links pedophiles with children. Roblox is that.
If being more open and public correlates strongly enough with "links pedophiles with children", then yeah, true. I expect Roblox to do plenty to improve its platform safety with their track record. The recent introduction of their ID verification system to prevent communication between users outside of specified age buckets, solely in the context of improving child safety, is working and significantly reducing cases of child exploitation both on and off of the platform.
> it facilitates pedophiles predating on children.
I don't agree that it facilitates this kind of behaviour, nor does there exist enough evidence to make such a claim. Try red-teaming it: take the place of a bad actor that aims to cause harm to a child on the Roblox platform.
First, the actor will need an account. Next, they'll need to join a game (one with a low content maturity rating) and find a vulnerable child. Of course, they can't actually communicate with each other, so the actor needs to verify their age. For this, they will need either the face of a child, a verifiable ID document of a child, or the ability to defraud the system (likely by forging an ID document).
Assume they can get past that, and are now considered in the same age bracket as the child, and can thus communicate with them. They now need to direct the child off-platform to avoid them being caught very easily by Roblox's safety teams and systems. They won't be able to use a social media link, as the actor isn't permitted to send them and the child isn't permitted to receive them due to them both being classified as minors by Roblox. So the actor will need to use chat or private messages to give them specific instructions, which will of course be heavily censored due to both parties being minors.
Okay, perhaps an easier way would be to have the actor create an experience with some unfiltered messaging system and let the child join them in it (if they don't have parental experience restrictions available, in that case the actor will be out of luck). They'll of course need to get the experience approved with a sufficient content maturity level. However, any unfiltered messaging system is going to be caught quickly by automated checking; all experience communications must go through TextService:FilterStringAsync(). Roblox's text-filtering is industry-leading, and much more protective than almost any other platform that allows free-form text communication.
Regardless, we'll assume that the actor is able to direct the user off-platform successfully somehow. In the case that a member of the safety team does follow up on any chat/message history and finds a Terms of Service violation, they'll request that the alternative platform take action, which will of course have no impact as the communications aren't against their Terms of Service.
> And if Roblox was IRL, it would've already been sued into oblivion
I hope I've effectively demonstrated that it's much more difficult to abuse or exploit users on the platform after recent updates. If Roblox was IRL, they wouldn't be able to implement any of these safety checks or features, and they probably would be sued into oblivion.
>If being more open and public correlates strongly enough with "links pedophiles with children", then yeah, true. I expect Roblox to do plenty to improve its platform safety with their track record. The recent introduction of their ID verification system to prevent communication between users outside of specified age buckets, solely in the context of improving child safety, is working and significantly reducing cases of child exploitation both on and off of the platform.
What are you, their PR agent?
>I don't agree that it facilitates this kind of behaviour.
It doesn't really matter if you agree that it does. It is a fact that it does.
To clarify, no <https://news.ycombinator.com/item?id=47334849>. I don't have any affiliation with Roblox Corporation, and I hope this will be able to make clear to you that my intention is to debate this matter in good faith.
> It doesn't really matter if you agree that it does. It is a fact that it does.
This isn't particularly constructive. Roblox publishes monthly updates on their safety status including statistics on the whole platform <https://about.roblox.com/newsroom>, and more in-depth statistics are also available for specific experiences <https://devforum.roblox.com/t/4028415>. There isn't any similarly concrete evidence to claim that Roblox directly facilitates links between child sexual abusers and children, so the difference between opinion and fact from that side of the argument is moot.
> There isn't any similarly concrete evidence to claim that Roblox directly facilitates links between child sexual abusers and children, so the difference between opinion and fact from that side of the argument is moot
No. Many legal cases were brought against Roblox due to these claims. Roblox implemented their new safety and age checks, and since then, none of these cases have progressed since because the evidence the provided by them for such is no longer relevant.
If there's any such evidence that I've missed, please do bring it to my attention by providing it.
People? There's a guy upthread quoting the Anthropic CEO on how they view the value of increasing training against the offset of the entirety of the $35T worldwide labor market... It's not "people". It's the salesmen.
>Regardless, I think if you are thinking purely from a ruthless business standpoint then standing up to the DoD was an incredibly ill-advised move.
It wasn't, there's been non-stop talk here for days about how Anthropic is a step-above, better-than-the-rest, the "only good AI" company. Enough already. It is a marketing tactic they are taking in opposition to OpenAI.
> So in theory (IANAL), investors can't easily bully Anthropic into abandoning their mission statement unless they can convince a court that Anthropic deliberately aimed to prioritize the cause over profit.
So why were they ever working with the military in the first instance, if that's the case? If you didn't gleam from OpenAI that it doesn't matter. Everyone is greedy and will jump ship for money if Anthropic does not get it for them.
reply