There is no excuse for submitting AI slop. When you file a court brief, you sign it indicating that you read it and it is as accurate as you can verify. You may quibble over details but you indicate everything you put in the file is factual.

Putting in fake case citations means you didn't read what you filed which means you violated your duty as a lawyer when you filed it.

Also - checking citations isn't hard. There's this tool called "Google" that you can spend 5 minutes with looking up citations. It doesn't need a law library - since all the case information is online. Takes maybe 5 minutes and something you can have your intern do.

Maybe an hour if you want to do a cursory glance at the case and make sure it's actually saying what you think it's saying. After all, nothing's worse than citing a case to say one thing when the case actually went the opposite way.

And honestly, I think the punishment could be simpler - you lose the case. Whatever it is. If a prosecutor did it and now causes a criminal to go free, well, lucky day for the criminal and the public will have their say at the voting booth for letting criminals go free. If it's a civil case, too bad, so sad, but now you have grounds for suing for inadequate representation.

Lawyers who lose their cases this way build a reputation and it's one where the free market and voters can easily resolve.

That's an exaggeration. You can give a LLM access to real things and they can use those real things to verify. I just flatly do not understand why they are not. It wouldn't make them infallible, but it would go a huge way towards improving the situation, and they are clearly not doing it. They could also use non-AI software tools to check up on the AI output. I'd bet that you could even use a plagiarism detection tool for this purpose with little to no modification, but I'd also bet this kind of tool already exists anyway.

It also shows that if the penalty is insufficient then they have no positive effect. A fine that people with a lot of money can easily afford is just a prohibition which only applies to the poor, with a license fee. Look to speeding tickets which scale with income for a fair model.

Yes

I'm not against a human doing it, I just don't get why they're not having the software do it, when it's feasible for it to do it.

What you're saying here is that no one should ever try to remedy the evils of the past.

That's stupid shit.

Fuck that stupid shit.

Debian used to just require old school Unix stuff plus Perl, which is pretty old school really.

Then they started requiring Python for some tools.

They just keep adding more requirements which bloat the base install. This is the opposite of the right direction.

The source code is the most important documentation about the source code.

Clearly this not even even being done by an automated tool, let alone a human. An LLM which is given access to a database of actual cases could reasonably be successful at checking whether the cased cited even exist which isn't being checked now!

Using non-local AI in any way in court filings which are supposed to be confidential until filed is glaring lack of professional conduct right up front. Allowing AI hallucinations to get in to your court paperwork even once is the same. They should lose their license for one year the first time, five years the second time, and permanently the third.

Slashdot is such a shithole

So this is Geniza from between the First and Second Temple? Sorry, just had to ask. :)

For an important API, yeah, it is probably a good idea if that API is something that you're told you can rely on, but I don't think this is still the case with the Linux kernel, where rust is more of a playground.

Except there are real drivers being written in Rust. It's being done because it eliminates a class of memory bugs that were tricky and difficult in C, and when you're dealing with complex devices, likely an overhead you don't want to deal with (e.g., GPU drivers).

Sure, if you're a hard core kernel developer, then you probably know the intricacies of the memory management. But if you're a slightly weaker developer trying to get hardware to work, well, you probably want some help so you can work more on driver bits and less on memory management bits.

Asahi Linux, for example relies on Rust on Linux code that's not in mainline yet - they're something like 600 patches that they have but cannot submit because the base dependencies are not in.

There was a brief spot between IE and Chome where Firefox had the market, but Google put that damn button on their search page that took everyone to a Chrome download and "wow"d people with the URL bar search.

I personally love Firefox, and for any minor problem it might have, I think the ability to have a reliable ad blocker without much hassle is well worth it.

Sure, let's ignore all the times during that heyday where Mozilla decided to alienate Firefox users. Sure, maybe they had a good reason to break the UI multiple times going away from XUL - first they get a new UI and then you needed an extension to fix it. Then they break it again and obsolete all the extensions you used. Users gave up and switched because it was a support nightmare where one day you start up Firefox and nothing works the way it used to because the update rolled out. Like what I needed to do that day was fix Firefox again because half my favorite extensions no longer work, or exist.

Firefox was on top and Chrome was the newcomer. Chrome did a lot of things better, but Firefox was still the king until they alienated users with this stuff that caused people to give up and switch. I mean, if I'm going to be burned by Firefox who decided one day I was going to lose basically everything, then I might as well check out the competition.

And now Firefox is where it is because they've refused to do things that users want - instead forcing Pocket and adware down our throats in the shadiest way possible. Like, do we NEED reasons to not use Firefox?

Earlier the entire Japanese localization team decided to quit. And likely taking the whole Japanese userbase with it because of the culture and the nature of the insult. In an era where Firefox should be able to pick up users easily it's still doing its best to shed them.

Crying over lost users while declaring "It's Google's fault! Monopoly!" when much of the damage was self-inflicted is not how it works. It even came back to bite them when the lawsuit threatened to cut Google's funding of Firefox.

Time to admit the damage was done, and then go about trying to attract users back. Maybe bringing back what was lost - why is changing the look still something I need to edit config files for - something we gave up in the 80s? Lots of Firefox customization is locked away in config files when it was a simple extensions to alter them.

There's a reason the Thunderbird team broke off because they didn't want to deal with the baggage Mozilla was bringing.

The 45 minute builds back in the 1990s .....

Obviously someone never tried compiling the Linux kernel back then. An hour to build was considered fast. It also was a good stability test because questionable computers would almost always crash.

These days the Linux kernel takes 5 minutes tops.

Android is also a beast to build - back in the early days, half a day to build it was common. Even on a high end machine you did a clean build in around an hour and a half. If you got a super tricked out Threadripper PC with SSDs you got it down to around 45 minutes. 64 core builds at the time were impressive. Of course these days we have 128 core PCs, but even Android 14 doubled the build time over Android 13.

Windows reportedly took 8 hours to build in the NT days.

In a little over 20 years we went from build times on things like Linux, GCC, Glibc, and other big projects which took the better part of an hour to just a few minutes. Fast enough that OpenEmbedded Linux builds everything from source - you set up a project and build it and it compiles the cross-compilers, the host libraries, and build tools and then spits out an image you can use in about half an hour.

Of course, the real thing is likely more WFH stuff - because if you walked in the door to the office, you were on the clock. At home, I suppose you could go through all that, but most people I know just close their laptops which puts them to sleep, so they just need to log into the VPN the next day. Hell, I'm super lazy, I just lock the PC and leave it running. It's not like the few watts the laptop consumes is going to kill me - I'm saving tons on gas and other things not going to the office so leaving the laptop plugged in and on isn't going to hurt matters.