![](https://lemmy.my-box.dev/pictrs/image/238bb7ef-8e89-44cb-9bc9-1560c98a9b3a.png)
![](https://b.thumbs.redditmedia.com/7figM5ltRdK7bAhIGenxdVFVRcQmMcW7PyYaCCKdCdI.png)
Because obviously when you train an LLM you include a large dataset of api keys for your new private service or the settings to generate them.
Because obviously when you train an LLM you include a large dataset of api keys for your new private service or the settings to generate them.
At the risk of committing whataboutism - the same can be said about Facebook, Google, etc. And I don’t say that to marginalise the problem. Quite the contrary - I honestly feel it should be illegal to gather this level of personal detail about people.
That the focus right now is on tiktok is suspect, to say the least, but I guess the upside is that the problem is getting attention now.
“But I use Jetbrains…” :(
I liked it best when it told me I had a very creative proposed solution for a specific issue.
It was very creative because it turned out to be impossible.
Your goldfish lived for 20 years?!
Yeah, because that worked so well with tracking popups.
As The Times told BleepingComputer last week, the attackers used exposed credentials to hack into the newspaper’s GitHub repos.
It explicitly says the credentials were leaked. If you’re really going to insist the word “hack” implies something else, I’m afraid you’re too far on the spectrum for me to continue this conversation. Cya!
For me it just gives an error post0 failed to load
.
Edit: that’s probably my lemmy client trying to approach this blog as though it were a lemmy instance 🤦♂️
Please point out where it states that Microsoft leaked it, rather than the more likely case of NYT leaking their credentials.
[Citation needed]
It’s unused, you can go ahead and kill it.
So… Unless Microsoft directly leaked those credentials, I don’t see how it would be their responsibility.
Truly the brightest timeline.
With these kind of titles, I hope you will, and know you won’t.
Moving goal posts.
Op said Don’t use AI, you’re saying Don’t use AI everywhere.
You first.
Stop? No. But results so far have shown a decrease.
Sure thing, bud. So far all the studies disagree with you, though.
Beats me, I don’t live in the US.
I stand corrected. It doesn’t include that as far as we know, on account of the bill not existing yet, not even in draft form. If you don’t mind, I’m going to ignore everything else you say now.
As someone who worked on a couple of video encoding / streaming services, this was an amazingly interesting read. Some personal highlights: