This is an indication of a systemic failure. And a human failure. Both.

But be thankful that you have the system that you have. In China, 99.97% of people arrested are convicted. The police use torture systematically to obtain confessions. Sometimes female suspects will be imprisoned with known rapists.

Be thankful that we're even having a discussion about systemic failure with the Flock database error.

You're inserting advocacy research (which is not reality seeking) into a case where the facts are already pretty clear.

About 20 are unarmed people.

About 8 of those unarmed people are actively attacking the police officer.

Of the remaining 12 -- some were escaping, some were attacking someone else, ..., and a proportion were genuinely innocent people.


This idea that "1/3" of people killed by strangers are killed by police is classic massaging the date, because almost every person who is killed or murdered is done so by someone they know.

The police are continually reforming. We need more reform. We also need a reality based discussion.

Data entry mistakes happen. What is worrying is that it wasn't corrected the first time she was pulled over. Also, the police should have issued an apology. It's the morally correct thing to do; however, I suspect they have legal advise specifically NOT to do this because of ambulance chasing lawyers. Nonetheless there is absolutely a systemic failure here that needs to be addressed.

A mistake was still made, in that the entry should have been removed from the database the first time.

PoliceViolenceReport -- an advocacy organization btw -- reports 2x fatal police incidents. So, since 2017, we have about 15 per year instead of 7. So you're way over your skiis. Did you read the "excellent data"?

And it isn't apples-to-apples. Because PoliceViolenceReport is measuring a different *thing*. We're no longer looking at just shootings, but other types of fatal incidents, such as the use of police vehicles.

The Washington Post reviewed shootings of unarmed individuals, and found 40% (!) were physically attacking the officer in question.

When it comes to HARMLESS people SHOT by policy, PoliceViolenceReport and the Washington Post agree.

That means that YOUR SOURCE agrees with the OP.

If there were really justice in this world

I get your point and I agree in principle.

$100M? Really?

Btw, if this were china, they have a 99.97% conviction rate. The police arrest someone. They are tortured until they confess. 3 in 1000 escape prison.

For all its flaws, be thankful that we live in a system that we have. A $100M lawsuit would not make it better, unless most of it went to pay defendants legal fees. We also need way more judges. Like 10x as many judges.

I suspect the Pareto Principle is very much at play here, in that 80% of customer issues can totally be answered by the scripts that human front-line call centre staff typically worked from (that's why the scripts exist in the first place, afterall), and therefore 80% of the calls can be dealt with perfectly well by an AI bot following the same script. It's only the other 20% that are the unique problems that KB articles and scripted/automated agents cannot handle, and - usually after a lot of pointless tick-boxing to get off-script - those are usually the ones that get escalated to a second line support peon that can actually take some initiative, and ultimately the few real corner cases to third-line support.

If AI can let you cull 80% of your support / call centre staff with no practical loss of service (according to the AI agent salesperson, anyway), then who isn't going to jump at that? Of course, they'll probably go for 90% or more and take extra hit on service levels, then have to wind things back a bit when they realise that the AI agents are not quite as good as promised and some people will get really stroppy about being forced to talk a bot, but at that point they'll have been paid their bonus and who cares, right? They're only customers, and you've got think of the shareholders!

Which is why, as many people pointed out at the time the root servers were opened up to all these new gTLDs, that this is just a license for the successful registrars running each gTLD to print money at the expense of everyone with a brand they have to protect. In most cases, they were absolutely right on the money; .biz being the classic example, almost immediately becoming a wretched hive of online scum and villainy most mail server operators equally rapidly blocked outright. (I am curious though; does anyone actually try and host a legit .biz domain these days? I don't think I've ever come across one.)

I'll give you, there are some pretty cool options for host/domain names under .web, especially with an arachnid theme or things like "web.web.web", that might become useful and popular for some groups, but apart from those - which Verisign, being Verisign, will no doubt have a special procurement process for to maximise their investment - the rest will almost certainly be parked/redirected brandname protection, a vast number of of scammer sites, and a handful of vanity projects that 99.99% of people probably won't even know about, and probably wouldn't care about if they did.

None of which changes the fundamental point that we seldom care about URLs any more. We click on links in apps, webpages, emails, etc., ideally having done some kind of due diligence to make sure the link is hopefully OK, or just cut and paste from the browser bar to put it into something we are writing. Functionally, "bankofamerica.com" is no different from "bankofamerica.web", or any other gTLD/ccTLD they might care to suffix it with, and it wouldn't make any difference to their service if all the other TLDs went away, unless they wanted to do some kind of content localisation based on a ccTLD rather than some other approach to achieve that.

Sam Altman is also a psychotic sociopath. He's probably pitching "Project Skynet" to Trump and Pete Hegseth right now - "Look how effective our tools clearly are! You could be using them against China, Iran, Cuba ..." [checks notes] "Um, Greenland, Canada, Democrats (especially Obama & Biden!), E. Jean Carroll, the BBC... And for the low, low price of..." [pinky finger] "one trillion dollars, we can be up and running by, let's say... 2029?"

I regret to inform you that you have woefully underestimated it. The actual retail rate offered to consumers is closer to $2200 US per gallon. Sources: internet-ink.com, cbc.ca. This $14 million fine is only worth like, seven thousand gallons, or less than 200 oil barrels of ink.

It's bad, but Microsoft has a awfully large number of lines of code to run through Mythos or whatever (whether they need that many is a whole other discussion). A more useful metric on overall code quality would be how many bugs are being found per 10k lines of code compared to their peers (including FLOSS); e.g. if Microsoft ran 10m SLOC through Mythos to get those 570 bugs, and a smaller project ran 1m SLOC and got 57 bugs, then you could reasonably argue that their code quality is about on a par with the smaller project. It's still Apples to Oranges though, because some coding solutions are going to be much more challenging to code, and therefore more likely to contain bugs.

On the upside, we're probably going to get several months of this while everyone with access to Mythos et al runs their existing code through it and integrates into their release processes for new code, and the end result will be things being much harder for all the bad actors in the world. Even if you don't use the improved code yourself, that's hopefully going to have a significant impact on the number and size of all the botnets out there, and that's a net benefit to everyone apart from the bad actors.

Wait, what? They're *making* money now? Last I heard they were still playing shell games with pretend money in a financial merry-go-round of pinky swear deals to make it seem like they are somehow not haemorrhaging quite so many hundreds of billions of dollars as they actually are to try and keep the VC funding flowing in.

The AI endgame, sure. That's totally going to be the kind of bait and switch that Google pulled when they transitioned from a search provider into an ad provider; get everyone hooked on your services, then monetise all the data you've captured and start cranking up the token fees until your customers (only they are actually more your "product" now) squeal, then turn it up some more while offering a rent-seeking subscription model that looks like a good deal until you realise (too late) what they've buried in the small print.

Drug pushers are probably looking at the tech industry in awe at this point.

Previous planetary probes have done figure-8s around the Earth & Moon to build up velocity before heading off for their target planets, so you could possibly do something similar with this to shorten the transit time; a few laps to provide initial acceleration to escape velocity, then coast to Mars using the panels to keep any systems ticking over and batteries topped off. Mars orbital insertion might need a little thought as to how to manage deceleration, or a secondary means of braking propulsion, if you can't do that using Mars' gravity alone though.

There are at least a few teams out there doing just that. In this case, finding software bugs, get one LLM to look for potential bugs, and use a second independent LLM to try and validate the potential exploits it finds / develop a PoC. Depending on what you are doing and how critical/sensitive the code is, you could also add more independent LLMs in each group to provide additional layers assurance before any output is passed over for human review.

There's also supposed to be a training loop with LLMs, so you should be flagging any false positives and feeding those back into your model so that the quality of findings improves. The current versions of Mythos/Fable might not be perfect, and probably never will be, but with a few more iterations Anthropic should be able to decrease the FP rate considerably, and ultimately that's going to be a big win for everyone with an interest in bug & exploit free code.