Now, knowing that the line between man and other animals is totally arbitrary, we should be fine to retell that story with beavers instead of people, right?

(I won't even get into how I flew there at 1000kph in a giant metal vehicle and carried a universal translator in my pocket, also neither of which were made by beavers.)

The model is that higher values of G/B are better, where G is good years per capita and B is bad years per capita, so the ethical thing to do is maximize G and/or minimize B. I called this "basic" because even though it tends to fall apart when scaled up, I think it captures what most people think of ethics. It's good for making tactical decisions about things like "should I punch that guy" or whatever.

For #1, where ethics would be just a fiction we made up to support a social game, what I mean is: People in a group have to play a prisoners' dilemma game where they find a balance of cooperation and sharing of resources. One way to reach that balance is to assume good will and treat others like you want to be treated. (Right up until someone violates that, then it's hyperbolic response time...) If ethics is just something people made up to get everyone to maintain that balance, then higher G/B being better is rooted in the same fiction.

For #2: My definition of G and B included good and bad experiences by animals, not just people. That means it's possible that over a billion years or so you could still achieve higher G/B by going back to a "natural" world inhabited by sentient but less intelligent animals.

For #3: It could be that the G/B in a natural world would actually be lower than if it was dominated by humans, because several billion people being able to live out their lives before the crash could result in a big enough G value to outweigh the eventual suffering.

For #4: If life as a human counted more towards G than the same amount of life as a beaver, then it could be a force multiplier for #3.

I say I'm on the fence because on a longer time scale, that kind of mental adaptation might not be that much different from genetic adaptation and we'll still have to find equilibrium or die out.

As far as the ethics question goes, if we use the most basic model and say that the ethical thing is that which promotes the most "good-life-years per capita" versus "bad-life-years per capita", where capita means one of the sentient residents of the planet, there are a few ways to look at it:

1. Ethics is just a fiction that humans have made up to support the prisoners' dilemma style of cooperation game we play, and there's no real answer.

2. The billon or two twilight years of our planet are better spent in rough equilibrium, where tigers exist, but most animals get some time to stretch in the sun, nuzzle their young, etc.

3. Life in equilibrium is actually cold, brutal, and short for most of the sentient animals involved, so providing a higher level of comfort for a population of ten billion people is better, even if it eventually crashes.

4. The experience of a human counts more than that of an animal, so a high human population with a decent standard of living for a long enough time pushes the needle past what a much longer natural equilibrium would achieve.

To be clear, I think it is. It's just not going to be a hologram of a wizard in a room you can ask a question to for a quarter, which is what these chat bots and copilots you see today are modeled around.

When I look at the kinds of AI projects I have visibility into, there's a parallel where the public are expecting a centralized, all knowing, general purpose AI, but what it's really going to look like is a graph of oddball AI agents tuned for different optimizations.

One node might be slow and expensive but able to infer intent from a document, but its input is filtered by a fast and cheap one that eliminates uninteresting content, and it could offload work to a domain-specific one that knows everything about URLs, for example. More like the network of small, specialized computers scattered around your car than a central know-it-all computer.

"Q20: Are Social Security numbers reused after a person dies?

A: No. We do not reassign a Social Security number (SSN) after the number holder's death."

https://www.ssa.gov/history/hfaq.html

They don't get reused.

There are about 70 years left.

The government is punting the problem to future people and there's no solution yet.

How many random people would you need to ask before you found someone who was pondering joining a gym this very morning and would love some more info about it?

The main reason they need to run in kernel mode is you can't do behavior monitoring hooks in user mode without making your security tool open to detection and evasion. For example, if your security tool wants to detect whenever a process calls ShellExecute, you can inject a DLL into the process that hooks the ShellExecute API, but malware can just check for that in its own process and either work around it or refuse to run. That means the hook needs to be in kernel mode, or the OS needs to provide instrumentation that allows third party code to monitor calls like that without running in kernel mode.

IMO, Windows (and probably any OS you're likely to encounter in the wild) could do better providing that kind of instrumentation. Windows and Office have made progress in the last several years with things like enabling monitoring of PowerShell and VBA script block execution, but it's not enough that solutions like CrowdStrike can do their thing without going low level.

Beyond that, there's also going to be a huge latency between when a security researcher finds a new technique for creating processes, doing persistence, or whatever and when the engineering team for an OS can update their instrumentation to support detecting it, so there's always going to be some need for a presence in kernel mode if you want up to date protection.