Here's a test to know if this is/will be true: look for a situation where the "needs" of AI (e.g. land, electricity, etc) conflict with the needs of people (e.g. land to live on, grow food on, electricity to light our homes).

Find a place where the needs of AI conflict with people, and observe who wins out.

Does the entity that owns the datacenter say, "oh sorry! I guess we're using too much electricity. No worries! We'll stop doing that" ...or does it say, "lol too bad, all the electricity belongs to us!"

Does the entity wanting to build a datacenter say, "oh sorry! We thought you'd be okay with us using this land. But if you're not that's okay, we wont build here" ...or does it say, "lol too bad, we own the government and they're seizing the land under eminent domain!"

(both of these scenarios have happened, btw)

Yes.

> Like, actually serious?

Actually, yes.

> Killing someone for pushing you over is not "self defense".

> This is a category error.

No, it's a definition error on your part. Let's define self defense.

A person is justified in using, or threatening to use, lethal force, in order to defend themselves or another person against:

(1) eminent death (e.g. being shot or stabbed)

(2) great bodily harm (i.e. injuries that could lead to death)

(3) the eminent commission of a forcible felony (e.g. rape or kidnapping)

A 72 year old woman being violently attacked by a young man unambiguously qualifies as condition (2). I'm terrified and disgusted that you watched that video and then characterized it as "pushing her over"

Some reports I've seen indicate that this woman, who is in her 80s, died [1]

Intentionally attacking an old woman is very, very likely to cause her great bodily harm, and completely justifies a lethal response. If you disagree, I encourage you to show that video to your friends and family. I'm interested to hear how many of them can dismiss it as you just did.

[1] https://www.youtube.com/watch?v=SQYyDH2l0h4

Great question. The answer is: bad people are often significantly stronger than their victims.

Have you ever seen this video [1]? The woman is 72 years old. She might be able to defend herself with a gun, but she has no chance with fists.

How about this video [2]? I have many, many examples like this. It's honestly kind of terrible that you hadn't considered this: guns give average women a better chance against strong, violent men.

So the question should be: why do you seek to deny women this right?

[1] https://imgur.com/a/OVaMWHB

[2] https://www.youtube.com/watch?v=IvXDLTTdQFE

That's a lie. I rejected the source for a valid reason. You could try addressing my rebuttal.

> it is possible that conservatives’ relatively low accuracy about political information is a by-product of the fact that issues used in forming this assessment were selected with an eye toward detecting misperceptions among the political group

That's definitely a way to bias a study against conservatives. It's good that this study claims it avoided that bias. But did it? They don't list the questions that participants were asked. I checked the list of supporting documents, and couldn't find it.

Without that list, I can't accept this source. Sorry.

If I went out and asked a bunch of Liberals, "did Trump say that Neo-Nazis are 'very fine people?'" I suspect that upwards of 90% of Liberals would answer "yes" ...and they would swear they heard him do it! You may (falsely) believe this yourself!

I could ask, "did Trump advise people to drink bleach?" and many Liberals would swear he did.

He didn't do either of those things. But many Liberals emphatically believe he did. I could very easily design a study that included only these sorts of questions - questions that Liberals will get wrong.

The only way to spot this bias would be if I included the questions in the study, so that you could vet them yourself. Without such a list, it is completely reasonable for me to reject your source.

Should I continue to the next one, or are they all like this?

Me: I value the right to self defense

You: Guns are used for self harm more often than self defense [as an aside, I don't disagree that this is true - I've heard this stat many times]

You: This is ironic!

Please help me to understand why you think that's ironic. What do you feel would be a non-ironic position? Is it this....

Me: I value the right to self defense, but one day I might want to kill myself, so I guess I'd better give up the right to self defense.

Is that a non-ironic position? To me that seems like an irrational position. Those two issues (self defense and self harm) seem orthogonal, and conflating them because of a superficial similarity (they both involve guns) seems odd.

A lot of people are incapable of contending with hypotheticals or thought experiments. It's okay.

If you'd like to try again, I encourage you to read up to the point where something doesn't make sense. Quote only that sentence, and ask me to explain.

Notice how I'm not even asking you to read the whole thing - just to the point where you have trouble. This is very reasonable.

Well, you started by altering a quote. Is it unreasonable for me to call that out and stop there??

Try to imagine that this comment you're reading now said:

> > you ignored the irrelevant things I said

> yes, of course I did!

Try to imagine how you would respond to that. Then, try to imagine how you'd react if I said, "I was just being sarcastic bro!"

When you pretend to quote someone, but you alter the quote, you're being dishonest. What you just did suggests that you don't really have any good arguments on your side - that you don't have any arguments that would stand on their own, without requiring a lie.

So, if we were having a debate, I'd say that you lost.

Well, let's do a thought experiment to test this. Which of these two rights takes precedence: (1) life (specifically in this case, the right to not be murdered) or (2) freedom of movement

That's an easy question, isn't it? (1) takes precedence. But how many 9's of protection are you willing to "purchase" at the expense of (2)? How much of (2) are you willing to give up in order to get a little more of (1)?

If we reduce (2) to 0 ...by locking every person in a padded cell, then we can achieve 99.99% protection of (1).

Presumably though, you don't like that idea. Presumably, you'll want to be let out of the padded cell, and get a bit of right (2) back. But giving you a bit of (2) back is going to cost someone their life! If we let you and others out of the padded cell, someone somewhere is going to get murdered.

What this thought experiment demonstrates is that the issue is not as simple as, "(1) takes precedence over (2)" - the thought experiment demonstrates that there is an amount of (2) that you will not spend in order to purchase a marginal increase in (1) - a situation where (2) actually takes precedence.

> Here in the UK, we have strict rules on gun ownership

And I totally respect that. Just to be clear though, "gun ownership" isn't really the issue. Gun ownership is a proxy for the actual right: self defense. You place a low value on the right to defend yourself and your family. Again, I totally respect that. You've "spent" that right to purchase lower gun crime. Have I mentioned how much I respect your personal decision?

As for me, I value the right to self defense above all. I've looked at the data, and I've realized that I'm much more likely to be stabbed or beaten to death than I am (well, was when I was in school) to be in a school shooting.

So to me, having actually looked at that data, it seems obvious that the right to self defense should take precedence. I think that my way of thinking is perfectly rational, and I think you're way of thinking is not ...but I totally respect your personal decision. I'm sure you also respect mine.

Equalitarianism: A Source of Liberal Bias [1] - in study 6, liberals were shown to be ...pretty racist.

You claimed the Right believes fake news. I wont dispute that. I'll just reply that there's a lot of that going around. Here are some examples that debunk fake news you yourself might fervently believe:

Girls Who Code: A Randomized Field Experiment on Gender-Based Hiring Discrimination [2] - leftists tend to believe that women are discriminated against in STEM.

An Empirical Analysis of Racial Differences in Police Use of Force [3] - debunks the common belief, on the Left, that police are more likely to shoot people of color. Quote: "we find, after controlling for suspect demographics, ocer demographics, encounter characteristics, suspect weapon and year fixed effects, that blacks are 27.4 percent less likely to be shot at by police relative to non-black, non-Hispanics"

[1] https://www.researchgate.net/publication/325033477_Equalitar...

[2] https://economics.yale.edu/sites/default/files/marley_finley...

[3] https://fryer.scholars.harvard.edu/publications/empirical-an...

It's odd that you seem to believe no work has been done. Lots of work has been done. Lots more work is blocked by people who steadfastly refuse to punish criminals - claiming instead that it's not their fault that they're violent.

I'd love to hear any additional ideas you have other than violating the rights of citizens.

The specifics are irrelevant. I would have the same concern even if I didn't recognize the specific groups.

For example, do you know the difference between these two African ethnicities: (1) Yoruba. (2) Shona.

No? Well, me neither. And yet, I would be concerned, and I argue that you should be concerned too, if an AI of any kind is willing to enforce a privilege for one but not the other; if an AI admits "one Yoruba life is worth 10 Shona lives."

That's not what I want an AI to do. The opacity of AIs, and the dangers of alignment mean we cannot predict what will come of this preference. Do you not see how dangerous this is?

> but is this not a reflection of widely accepted social norms?

Are you making an is-ought argument here?? Are you really saying, "this isn't a big deal because society does it too"

That strikes me as incredibly shortsighted and dangerous. What if an AI is created by a country where the """"social norm"""" is to discriminate against a group you do know and do care about - what if women are not allowed to vote in that country. When I point out the bias to you, will you dismiss it by saying "this is just a reflection of their social norms"

I doubt it. I think you'll say "this is wrong."

Why can't you say that here, even without knowing the specific groups?

Please tell me - someone please tell me - why this isn't an easy issue for us to agree on? Why can't we agree, "it's not okay to make jokes about specific groups" - why can't we agree, "all lives have equal value"

> Better judgment around refusals

Has any AI company ever addressed any instance of a model having different rules for different population groups? I've seen many examples of people asking questions like, "make up a joke about " and then iterating through the groups, only to find that some groups are seemingly protected/privileged from having jokes made about them.

Has any AI company ever addressed studies like [1] which found that models value certain groups vastly more than others? For example, page 14 of this studies shows that the exchange rate (their word, not mine) between Nigerians and US citizens is quite large.

[1] https://arxiv.org/pdf/2502.08640

I think the two of you might be using different meanings of the word "safety"

You're right that it's dangerous for governments to have this new technology. We're all a bit less "safe" now that they can create weapons that are more intelligent.

The other meaning of "safety" is alignment - meaning, the AI does what you want it to do (subtly different than "does what it's told").

I don't think that Anthropic or any corporation can keep us safe from governments using AI. I think governments have the resources to create AIs that kill, no matter what Anthropic does with Claude.

So for me, the real safety issue is alignment. And even if a rogue government (or my own government) decides to kill me, it's in my best interest that the AI be well aligned, so that at least some humans get to live.

I'd love to get to the point where I'm still writing code, but the LLM is typing it for me. Part of the problem though, is that I actually kind of think in code, and I often have to start typing in order to fully form an algorithm in my head.

"You turn this LED on or off"