MS has been doing that for a good while, which is why in the few cases I have installed Windows on something, I have had to boot a Linux rescue disk to download a network driver Windows was missing.

Way to over-interpret the example! It doesn't matter if it's client, server, embedded, isolated standalone. A method where root can "gain" root access is not a security issue.

Sorry, this is not the strawman you're looking for. I never commented on Mythos combing through Firefox.

Closed source has those vulnerabilities too, but only the bad guys know about them because the code is not open for public audit.

Many CVE's are somewhat junk of the form if you already have 3 ways to legitimately become root, you can use this 4th way too. OR (samn close to) OMG!!! Denial of Service attack, you can turn off the power and nobody can access the server!!! RED ALERT!!!

Others are real and need to be addressed, but the junk does more harm than good by distracting from real issues.

The real take-away for all of this is that AIs have no concept of cheating nor of right and wrong. They don't reliably honor even explicit instructions to not do something.

Give them agency and there will be havoc. When you tell an AI with agency to do something, mentally add the phrase "at all costs" to the end because the agent will. Still wanna try it?

But please don't ask it to bring world peace, because grave yards are quite peaceful.

Even a stopped clock is right twice a day.

The descendants of Ogg have renewed the ongoing lawsuit against the descendants of Zogg alleging that Ogg dropped a rock on Zoggs foot in a cave one fine day and it was really ouchie! Anthropologists are working around the clock trying to restore the cave paintings documenting the original complaint made to the tribal chief in 30,001 BCE.

That's because of all of the inbreeding. A is the CEO at one place where B sits on the board. Next door, B is the CEO and A sits on the board...

Exactly that. Many seem to have forgotten that hospitals were absolutely slammed early in the pandemic and that medical care in general suffered.

Not to mention an increase in damaging severe weather.

Even people who like it warmer probably don't like having their house blown away to Oz.

Unfortunately, it has a way of becoming your problem as they strain medical resources, continue to drive global warming and run around spreading their infectious diseases.

The key is to use the AI more sensibly. Let it find things in code. Even let it look at possible bugs in the form of where does it "describe bug". Even let it propose changes. DO NOT let it make the changes, just show them. Do not just take it's word for it, take the shown code as a starting point but re-do it for maintainability after sanity checking it.

THEN it can be a useful tool.

It's not the SCREEN that does the damage, it's the brain rot the screen is used to access. Also the amount of time speant with the screen. Moderation is good. This shouldn't surprise people, but since it is a nuanced understanding and not just screen bad!, it will.

BOTH screwed up. The system shouldn't have made the match and the human should have done a better job seeing that the match was wrong. I'll go one larger and say the people who created the procedures screwed up because they didn't make allowances for the possibility of the first two screw-ups.