#167: AGI: Where Will it End? (Bret Weinstein & Heather Heying DarkHorse Livestream)
In this 167th in a series of live discussions with Bret Weinstein and Heather Heying (both PhDs in Biology), we discuss the state of the world through an evolutionary lens. It’s our three-year anniversary doing these livestreams!This week we revisit raccoon dogs and the origins of SARS-CoV2. Then: Artificial Intelligence and ChatGPT—should we be worried? If so, how worried? Bret proposes three categories of AI: malevolent, misaligned, and deranging. Finally: trans ideology and activism: trans...
We've come into you with another prime numbered episode, and it is our third year anniversary of doing these live streams.
And we're going to say just a little bit more about that after we get through all the top-of-the-air logistics, but we're going to talk to you a little bit about today.
We're going to come back with a slight update on the raccoon dog story we talked about last week.
The raccoon dogs are unaltered, but the story has moved ahead a bit.
Yes.
Um, yes.
As far as I know, they're unaltered.
I don't know.
I didn't interview any of them.
Nor did I do the animal behavior on them.
I actually did see a, uh, I don't know what the origin of it was, but somebody did place online a brief interview with a raccoon dog who, um... Was it Brendan?
It was a bizarre interview in which the dog did make the correct argument that, in fact, it was slated to be eaten and therefore was entitled to spread whatever it wanted.
Um, but anyway, I have digressed.
Is this a complete farce, or did you actually see something?
I am reporting.
I am reporting.
I am not compelled that this was an actual raccoon dog, nor that the words were, um, in fact, raccoon dog in origin, but I did see something to this effect.
Okay.
I have no idea what's happening.
So after that, we're going to talk a bit about AI.
Brett has a number of things to say about AI, going back a long ways, right?
And then we're also going to talk about some things in the trans universe that have transpired this week.
And that, that will be it.
But let us, let us get to the top of the hour stuff first.
We follow these live streams with a live Q&A.
You can ask questions at darkhorsesubmissions.com.
If you're watching live, you can join the chat at ODYSSEE.
No, is it two S's?
I know it's not spelled the way Odyssey is spelled, but I'm... It's O-D-Y-S-E-E, okay.
You please consider joining me at my sub stack, Natural Selections.
Last week I wrote How Now, Cal of Brown, and this week it was Stark and Exposed, a slight adventure into modern agri...
Not agriculture, not at all.
Modern architecture and the ways in which the movement that included Le Corbusier and Gropius and such actually very much mirrors and was part of the postmodernism that took over so many domains in the mid-20th century.
So that's in Natural Selections this week, which I encourage you to check out.
We have, of course, a store which is run by, and the print shop is also owned by, a couple who are just doing an amazing job.
And Zach, do you happen to know, because I've taken it out of my notes here, what the URL for the store is?
Is it just Dark Horse Store?
There.
DarkhorseStore.org.
So consider checking things out there.
We've got things like stickers, which you'll see on the back of my laptop right there.
Darkhorse stickers.
And soon, or maybe already, pins.
Lapel pins.
Yes, and not lapel, but you know, anywhere you want to wear them, really.
I mean, I suppose.
If you are someone who occasionally needs to be shaken vigorously to bring you back to your senses, then you should wear lapels.
Otherwise, it's pretty much up to you.
But the question is, what are pins?
Well, if you wear them, which you could do because sometimes you need to be jarred back into reality, or you could just wear them for stylistic reasons, then the pins would go well on them.
I remember when we were professors, and one of our most wonderful and most frustrating students, who was a very tall and broad man, much larger than me, was just not doing the work that I knew he was going to do excellently if he would ever turn it in.
And you said to me, you just need to advise him to wear lapels to class the next time, and then you can reach up and shake him by them.
Yes.
And I don't think it worked.
I believe I know exactly who we're talking about, and I believe that later on in our trajectory through that strange space, he came to class wearing a t-shirt and a separate set of lapels that he had... He had had fashion just for the occasion, knowing that sometimes people would want to grab him by the lapels, even though he towered over almost everyone.
He had no illusions about himself, and he understood that he did need to be shaken by his lapels every now and again, and wished to facilitate, you know?
It's an honor to be respected enough that somebody would put on lapels in case you needed to shake them.
Yes, indeed.
Okay, and of course we have Hunter Gatherer's Guide to the 21st Century, which not only is it available everywhere books are sold, but there is also signed copies available right here on the San Juan Islands at Darvel's Bookstore on Orcus, which you can actually go and get signed copies.
We just signed a bunch more this week, or you can order from them.
That's Darvel's.
And we are supported by you, of course.
We appreciate you subscribing, liking, sharing both our full videos and the clips that are produced on both YouTube and Odyssey.
And of course, we've got the full audio episodes everywhere.
everywhere that you might listen to audio.
You can also support us by joining us at our Patreons.
Tomorrow at my Patreon we have our monthly private Q&A, so if you join us there, it's a smaller group, we actually interact with the chat, it's really a ton of fun, and you also have monthly conversations at yours, so we encourage you to join us there.
You can access the Discord server where they have book clubs and karaoke and all sorts of fun things, and of course we have We are very, very grateful for our sponsors.
We pick and choose them carefully, and we presume that they pick and choose us carefully as well.
And so, as always, we start these episodes with three ads.
Without further ado, here we go.
I managed to find the ad in time.
I did hand it to you just moments ago.
Oh, you did.
No, and I had already buried it.
I was able to find it, in part because my feet are so comfortable and well taken care of that I wasn't distracted by any pains.
That did help.
Yeah, it helped.
And what do you attribute the awesomeness of your feet?
I attribute it almost entirely to the fine folks at Vivo Barefoot Shoes.
What a coincidence.
What a coincidence.
They are our first sponsor and they make shoes for feet.
Everyone should try them and it is amazing.
Most shoes are made for someone's idea of feet.
Vivos, however, are made by people who have feet and know how to use them.
And word is spreading.
We have been asked by strangers if the shoes we are wearing are as good as they have heard, and yes they are.
Here at Dark Horse, we love these shoes.
They are beyond comfortable.
The tactile feedback from the surfaces you are walking on is amazing and they cause no pain at all because there are no pressure points forcing your feet into odd positions which might cause you to misplace things and not be able to find them because you were, for example, distracted.
They're fantastic.
Our feet are the product of millions of years of evolution.
Humans have evolved to walk, move, and run barefoot.
Modern shoes that are overly cushioned and strangely shaped have negatively impacted foot function and are contributing to a health and disorganization crisis because people move less than they might, in part because their shoes make their feet hurt.
Enter Vivo Barefoot.
Vivo Barefoot shoes are designed wide to provide natural stability, thin to enable you to feel more, and flexible to help you build natural strength from the ground up.
Foot strength increases by 60% in a matter of months just by walking around in them and organization improves by 70%.
That is a personal anecdote, but I'm 70% better organized since wearing Vivo barefoot shoes.
Dude, you need to start wearing them to sleep.
I need the other 30%.
Yeah.
No, no.
I just think that you still have a long way to go.
There's an argument to be made.
Maybe my gloves are too tight.
That could be.
The number of people wearing Vivo barefoots is growing.
Once people start wearing these shoes, they don't seem to stop.
That's something that's not stopping.
I just gotta get through another paragraph or two and we'll be fine.
I mean, the fact is that this, like, Vivo is one of our oldest, our most, our Our oldest sponsors, that sounds wrong.
But we love these shoes, and I think that they can accommodate any amount of silliness in the ad read.
Yes, I think they have.
But you're attributing your still abominable, but you're arguing slightly improved organization to wearing the shoes.
70% better.
You just don't realize it because that 30% is still a bummer.
I'm not sure that making up data during an ad read is actually a good look.
I labeled it as an anecdote, but I will say It's 70% better, which still doesn't leave me in a good spot.
Can you operationalize organizational improvement for me?
Yes, let's do that after the podcast.
Vivo Barefoot has a great range of footwear for kids and adults for every activity from hiking to training and everyday wear.
They're a certified B Corp that is pioneering regenerative business principles.
Their footwear is produced using sustainably sourced natural and recycled materials with the aim to protect the natural world so you can run wild on it.
Go to vivobarefoot.com and use the code DarkHorse15 to get an exclusive 15% off.
Additionally, all new customers get a 100-day free trial so you can see if you love them as much as we do.
That's v-i-v-o-b-a-r-e-f-o-o-t dot com and use the code DarkHorse15 at checkout.
Good job, man.
Thank you.
Our second sponsor this week is Helix.
Now, Helix makes fantastic mattresses, and they might not want you to wear your Vivos to bed, so you're going to have to sort of do a trade-off.
Do you think that the Helix mattresses also improve organization, or just the Vivos?
I don't know.
It's for you to decide and produce your own anecdotes next week.
My feeling is if there's a trial period, one should not wear shoes on the mattress during the trial period, but afterwards, who's to say?
Who's to say, yes.
Helix makes fantastic mattresses that are supremely sleep-enhancing.
It's amazing what a difference an excellent mattress can make for good sleep.
Our animals love it, too.
A little too much.
Maddie, our Labrador who has left the building, is not allowed on the bed, and she knows it.
But we know that she takes naps on the new Helix mattress when we're not around.
It's just that good.
Seriously, Helix Sleep is a premium mattress brand that offers 14 different mattresses based on your unique sleep preferences.
Take the Helix Sleep quiz online and in less than two minutes you'll be directed to which of their many mattresses is best for you.
Do you sleep on your back, your side, your stomach?
Do you toss and turn or sleep like a log?
Do you prefer a firmer or softer mattress?
All of these are taken into consideration with the Helix Sleep quiz.
Sleep like a log has never quite struck me as right.
Yeah, logs don't sleep.
Logs don't sleep.
Logs don't sleep.
They just kind of lay there until they don't, at which point they're not laying there like logs either.
Okay.
Logs don't sleep, but dogs aren't especially sick and clams aren't especially happy, as far as I can tell.
So there's a lot of wiggle room.
Is it that they appear to, like, if you look at them just right, if you turn it just right, it appears that they're smiling, sort of like Cheshire Clams?
I don't think so.
I think it's people projecting.
Happy people, apparently.
Happy people, sick dog people, log sleeping people.
Right.
However, your preferences in sleep are all taken into consideration with the Helix Sleep Quiz.
Once you've found your perfect mattress, it ships straight to your door, free of charge.
Then you'll have 100 nights to try it out with any penalty, during which you should not sleep in your shoes.
If you love it, which you almost certainly will, they've got 10 to 15 year warranties on all of their mattresses.
Helix has models with memory foam layers to provide optimal pressure relief if you sleep on your side, and models with a more responsive foam, which provides optimal support for stomach and back sleepers.
Enhanced cooling features keep you from overheating at night, and every Helix mattress combines individually wrapped steel coils in the base with premium foam layers on top, providing excellent support for your spine and comfort for all of you.
Helix has been awarded the rank of number one mattress by both GQ and Wired magazine, and we can see why.
Helix mattresses are made in America at their very own manufacturing facility.
They're built for human bodies and built to last.
Helix also supports military, first responders, teachers, and students by giving them a special discount on site.
We look forward to our Helix mattress providing us with years of excellent sleep.
You should look forward to yours doing so as well.
Helix is offering up to 20% off all mattress orders and two free pillows for our listeners.
Go to HelixSleep.com slash Dark Horse and use promo code Dark Horse.
That's HelixSleep, H-E-L-I-X S-L-E-E-P dot com slash Dark Horse.
This is their best offer yet and won't last long.
With Helix, better sleep starts.
Two points of clarification.
One, the sleep quiz should be taken while awake, but not driving.
Two, when they say sleep on your side, they mean your orientation rather than sleeping on your side versus the other person's side, which would make it your side.
Yep.
It's gonna be one of those kind of days.
I can tell.
Yes, you can.
After this many years, you should be able to.
Our final sponsor this week is Maddie's Favorite.
I mean, she likes the shoes just fine, because sometimes when we put them on that means we're going out for a walk.
She likes the mattress a bit too much, but her favorite is Sundaes.
Sundaes is dry dog food.
When they approached us, when Sundaes approached us about being a sponsor, we were dubious.
Maddie, our dog, is absent at the moment.
is Labrador.
Frankly, I was going to... Brett did an ad recently for a guest episode in which we staged it so that you would feed her during the ad, and it was awesome, and I was going to do that, but we ran out of Sundays last night, and Maddie was like, I am not showing up on camera if you don't have Sundays for me.
No.
We'll see.
We'll see.
But she's a Labrador.
Labs will eat basically anything.
What possible difference, we thought, was she going to show an interest between her usual kibble, a widely available high-end brand, and sundaes?
Well, we were wrong.
Maddie loves the food the sundaes makes.
Seriously, loves it.
And indeed, when we ran out and gave her some of that other high-end brand this morning, she looked at her bowl, looked at us, sat her butt down, and didn't eat.
She wanted her sundaes.
Guess what?
It is far better for her than the standard burnt kibble that comprises most dried dog food, even the high-end stuff.
Sundaes is the first and only human-grade air-dried dog food.
Combining the nutrition and taste of all natural human-grade foods with the ease of a zero-prep ready-to-eat formula, Sundaes is an amazing way to feed your dog, and in a pinch, A person, too.
Yep.
I didn't try to feed you with it.
You tried to feed you with it.
Best dog food I ever tried.
Sundaes is... Sundaes is easy for humans, too.
No fridge, no prep, no cleanup, no gross wet dog food smells.
And you may be able to feed your husband on it.
Sundaes is gently air dried and ready to eat.
I mean, you felt I felt, oh yeah, it's fine dog food, but you know.
Trying a taco?
Got a little crunch?
Going a bit far.
Okay.
Yep.
Sundae's is gently air-dried and ready to eat.
No artificial binders, synthetic additives, or general garbage.
Seriously, look at the label.
All of Sundae's ingredients are easy to pronounce, except for quinoa, and healthy for dogs to eat.
In a blind taste test, Sundae's outperformed leading competitors 40 to 0.
I thought that was going to be a made-up number when I first saw that too, but again, here's the thing.
When I have a bowl of her previous food, which we still have on hand, Ready for her.
She's enthusiastic, usually, but once she's been having Sundays, she's not enthusiastic, and she always goes for the Sundays.
Preferentially.
Always.
She bounces, and she spins, and she does all these amazing anticipatory Gymnastic leaps in anticipation for the Sundays and not for the other stuff.
Do you want to make your dog happy with your diet and keep her healthy?
Try Sundays.
We've got a special deal for our listeners.
Receive 35% off your first order.
Go to SundaysForDogs.com slash Dark Horse or use code Dark Horse at checkout.
That's S-U-N-D-A-Y-S F-O-R-D-O-G-S dot com forward slash Dark Horse.
Switch to Sundays and feel good about what you are feeding your dog.
All right, question for you.
Okay.
Why is quinoa spelled that way, given that the Inca did not have written language?
It should just be spelled phonetically, right?
It's, you know, we actually know the person to ask.
Yes, we do.
We have a Quechua expert in our sphere.
I mean, everyone does, but... Yeah, so, it's always a question when languages that weren't written at all or weren't written in our alphabet are turned into being spelled into our alphabet.
Often it feels like, why'd they spell them that way?
And sometimes it's going to be about the way the first person who was being interviewed about their words was pronouncing it, or the hearing on the other end, or sometimes I feel like it's just intentionally befuddling.
Ellis Island phenomenon.
Yeah, yeah, exactly.
But yeah, Quinoa does not seem very much like Quinoa.
Quinoa and Quinoa are miles apart.
Yeah.
Kilometers apart.
All right.
So We've been at this for three years.
Yes.
Three years, and... Feels like two of those years were this morning already.
Sometimes, yes.
Sometimes.
Our very first one was on March 24, 2020, at which point the lockdowns were beginning.
It was the wild, wild west out there, and I don't think we did or could have predicted half of what has happened since then.
But we are, as ever, grateful to our audience and to the remarkable number of people who get in contact with us.
Yeah, none of that was ad.
Yeah, even the discussion about quinoa was totally unremunerated.
Our producer has just returned from the other side of the border, so he's still... there's no excuse.
I said border.
I meant equator, actually.
Lots of borders.
He crossed lots of borders, only one equator, and now I'm thrown again.
I can't even get it right.
Equator, border, one equator crossing, multiple border crossings.
He's back.
That wasn't an ad.
This is the third anniversary episode of our Dark Horse live streams.
You were doing Dark Horse for A bit about three, six months before that, but we've been doing these weekly.
We started off doing them twice weekly, pretty much reliably ever since, and it has been quite a wild Yeah, and I should say, Dark Horse started as a discussion program, me and one other person, and we were doing it from a studio in downtown Portland.
When COVID happened, Zach and I grabbed everything we could pull out of that studio and moved it to our place and got a lot of stuff from a hardware store and built the set that we were on the whole time we were in Portland under lockdown.
But that the idea for the live streams came from you.
What you said was we were doing something, trying to sort out what was going on with COVID, you know, trying to use our biological toolkit to make sense of it.
And you felt that we really needed to share what we were doing because everybody was struggling to figure it out and it might be useful and that was a tremendously good idea and we know because we run into people all the time who tell us what this meant to them during the COVID period that it was the right idea and I'm so glad that we did it in spite of all of the terrible stuff that came back at us in many regards.
Yeah, no, it's been clarifying again.
We had an evergreen moment of scales falling from our eyes and finding clarity around both issues and relationships that could be painful, but as I began saying back then, I'm glad I know.
Yep.
And once again, that's the case.
That's the case throughout COVID as well, that as painful as it can be to realize that people, that the relationships, some of the relationships you have aren't what they seemed, and people don't have the values that they claim to have.
I'd rather know.
I'd always rather know.
And that's, I mean, I think one of the things that you learn in all of this is that some people wouldn't rather know.
Right?
Some people would rather stay in the dark.
And of course, just as when Evergreen blew up, that's coming up on six years now.
We, yeah, lost some relationships but gained so many more that were extraordinary and we have made some just amazing both professional relationships and friendships since we've been doing these live streams and are very grateful, very grateful for that.
It's better than that.
Awesome!
Yes.
Is that the first time you ever said that?
Not in life, but... It's even better than that.
It's even better than that.
Terrific, let's do it.
So, you said you're grateful to know.
Yeah.
And I would go way past that.
It is essential that you find out who it is who can be relied upon under crisis circumstances, because the worst thing that can happen is you can imagine, based on how people behave under normal circumstances, What they're going to do in a crisis, you can be dead wrong, and then when the crisis happens, the cost of discovering that in real time is spectacular.
So as terrible as it was to have seen people that we would have imagined would stand up under crisis conditions fail in that, as you say, it's better to discover it so that you don't put weight on that ice.
And the people who you discover in the crisis, the people who shine.
Sometimes people surprise you, or other times you meet people that you never would have encountered absent the crisis, and you discover that they are up to the challenge.
It is admittedly painful, but it is an upgrade to the quality of life.
Yes.
So anyway, the universe, when it puts you through a crisis and reveals you can't handle it, is doing you a favor.
And it is important to realize that that's what's happening, as much as it is heartbreaking to lose people.
Absolutely.
So, thank you.
Indeed.
Okay, and just one brief comment about a story we talked about at some length last week, before you start talking about AI, is the raccoon dog research that we discussed last week.
And we read a number of bits from the Atlantic article that had come out and shared some of the New York Times.
And part of our objection was, just like with The TOGETHER trial.
The results were released and promoted in mainstream media as if they were scientific findings that everyone could assess, and therefore the fact that they were out there sort of implies that they have been assessed by the scientific community.
But there was literally no paper at all.
There was a press release, I think, or maybe just a couple of interviews.
There was nothing.
Well, that has now changed, kind of.
So, on March 20th, just, you know, five days ago, a couple days after our last livestream, they did indu- indude.
I'm trying, that must mean something, but I don't know.
They did indeed post something to Zenodo, a preprint server, although they say, they call it a report, and they say, quote, I actually don't have the quote, they do not intend to publish this report in a scientific journal.
Really?
I wonder why.
Well, when you look at the paper, you can begin to kind of see why.
I have only skimmed it.
I have noticed that a lot of other people have already dug deep and found, you know, various problems with it.
It's messy and confusing and pretty long, and the authors say in their little preamble they're not going to publish this as a scientific paper.
Like, what, does that just give them cover?
Like, oh, we put a thing out, but, well, it's not really the real thing.
So, you know, it's just like, it's one feint after another.
And near the end of the paper, they say the authors claim that quote, data accumulated since the beginning of the COVID-19 pandemic point clearly to a zoonotic origin of SARS-CoV-2.
And then they proceed to make a series of the weakest arguments I can possibly imagine.
I mean, it's really remarkable what like the level of analysis and that they basically don't expect anyone to check their work.
Like, A, they started by not letting anyone check their work, and now they put something out where they're like, yeah, but it's not really a paper.
We're not going to submit this to a journal.
And yet they make these claims that just sound like the strongest claims possible, but the actual claims aren't.
They're insane.
Well, in fact they said this was the strongest evidence yet for a zoonotic origin.
This, this like finding some, some bits of raccoon dog genome in the same place as some SARS-CoV-2.
Right, some really weak circumstantial evidence that is actually perfectly consistent with the idea that the Wuhan Institute of Virology is close enough to the seafood market that people who work at one are liable to shop at the other.
And we know there was SARS-CoV-2 there, and we know there are raccoon dogs there, so this evidence doesn't give us anything to do, even.
It's not evidence.
And so, what they are doing, and we need a name for this, and at the moment I don't have one.
It's somehow...
It's like headline priming or headline ginning, right?
The whole idea is to pretend that there is science that justifies a headline and that it's a data flow problem where, you know, they're gonna put out the conclusion of the paper and of course the paper's coming and it will, you know, it's basically The bad habit that science folks get into where they will read the abstract of a paper and they will not figure out whether the method section justifies it or the statistics that were done were correct or any of those things.
So the idea is you read the abstract and the whole thing is based on trust.
This is like, well we're not even going to give you an abstract.
What we're going to give you is the headline you would write if there was an abstract that was justified by a paper and then we're not going to follow up with the paper because of course The value of the whole thing is the exercise of producing the headline, and why would you go through the, you know, why would you put yourself at risk of having your headline invalidated by scrutiny when you could just allow the headline to float for good as if a paper had emerged?
Yes.
This has been going on a long time, by the way.
Yes, it has.
I remember this from the death of Dolly the sheep.
So this is going to have been like late 90s?
Early aughts?
This is going to be early aughts and I had a dog in the fight because in my paper on telomeres I had predicted that she would die of age-related pathologies anomalously early and They said she didn't.
Did they give us a paper?
No.
They said, they literally put out a press release in which they said that they would do a full necropsy and they would let us know if they found anything.
Right?
Which is insane because at the time that Dolly died she was of course the oldest mammal clone on earth.
And so a full necropsy doesn't begin to explain what they should have done for this animal.
A, they gave the implausible explanation that they had put her out of her misery because they couldn't stand to see her suffer from the pathology that they swore was not age-related.
But, you know, okay, whatever the cause of death, this animal should have been thin-sectioned, one end to the other, so that future researchers could go back and figure out what was taking place there.
Necropsy It's not the beginning of what they should have done.
But anyway, my basic point was, there was no paper there either, right?
This was a case in which there was a lot at stake, a lot financially at stake, because the patent on nuclear transfer cloning was incredibly valuable, and its value would be lower if it caused age-related pathologies.
So in any case, the trick of creating a headline or creating an impression that scientific work has been reported, but not actually presenting the work,
is a tactic that one should expect to see where there's a great deal at stake either with respect to something like COVID origins or with respect to something like a valuable patent like nuclear transfer cloning and I would also point out that interestingly that report on raccoon dogs it came in advance of the revelation of a paper that Robert Malone called smoking gun
On COVID Origins, where basically it reported an experiment in which... This week?
In the last four days?
Last week, somehow.
But I mean, you just said that this, the report here, predated... The headline, the initial report in which we all got the word that something important had happened with raccoon dogs, right, preceded by less than a week the revelation of this published work in which the enhancement of Sars-like coves to become infective of humans was discovered in the literature.
So anyway, check out Robert Malone's Twitter and search for the term smoking gun and you will find that research.
I'm reminded, you can show my screen just briefly, this is an article published in Forbes in 2020, in July 2020, in which they say you must not do your own research when it comes to science.
And they make this argument that it's just too tough, it's too challenging, and you're putting yourself and everyone else at risk by doing so.
And compare that to, Zach, if you would give me my screen back for a moment so I can show this paper, you can now bring up this.
This is the report that they say is not destined for a journal, in which they make introductory remarks and say, among other things, Oh, they say somewhere in here that they are not going to submit this to a journal, but this paper is... Oh, this report is not intended for publication in a journal.
There it is.
They start with key points, which is sometimes done in an actual scientific paper, but this is accessible If they would share more information, the likes of which you're supposed to do in science.
And to have this kind of analysis happening at the same time as, granted that Forbes article is from almost three years ago, but throughout this we've been told Don't ask questions.
Don't do your own research.
Don't assess for yourself.
Simply trust the authorities who've shown up with credentials that you may or may not be able to assess, with backgrounds and with skills that you may or may not be able to assess.
And don't, you know, pay no attention to the man behind the curtain.
Pay no attention to the fact that the entire system has been so, so corrupted by the role of money and federal granting agencies That the vast majority of people who are now doing science in the West, at least in the US, are doing little tiny reductionist pieces of science and will actually say if you ask them questions outside of their little domain, well that's not my concern.
So why would we trust You know, world health to people who will say, no, I just do this thing.
I just do this little thing.
And when you go looking, I spent a little bit of time looking at the backgrounds of a few of these authors who, you know, aren't the ones that we keep on hearing about.
And indeed, they're just doing this narrow, you know, blinders on science for their entire careers.
And where's the evidence that they actually know how to assess things?
I don't find it.
No, and clearly they don't for them to put their name to this.
Clearly, they don't uphold any such standard.
And I will point out, so much of this, so much of the malfeasance here come down to the application of two different standards depending upon whether you're on-narrative or off-narrative, right?
Almost all of this, the things that people who are dissidents are accused of violating the sacred rules of science, The fact is, those who are on narrative are granted no end of leeway in the same direction.
So if you were to simply apply any consistent standard, you would come out with a far better answer than we have.
And in fact, what's happening is, you know, somebody is peeking at who it is who is engaged in work and deciding whether to institute an impossible standard that can't be met, or no standard at all, so that you will meet it Irrespective of whether what you've effectively done is publish a newspaper headline without a method.
Yep.
And anyway, of course this crashes the world, right?
Science is not capable of, you know, science isn't a posture.
It's not a costume, right?
It's also not optional.
Right.
And we've taken it for optional.
It's like, it's an affectation for many people who are calling themselves scientists.
Yeah, you actually have to do the method correctly and you have to do it according to the underlying philosophy of science or it doesn't work.
Yeah.
Right?
The fact that it involves science-y looking stuff or science, science-y language has nothing to do with it.
People with degrees, it doesn't matter.
You either did the method right or you didn't.
Yeah.
As long as we spend a little bit of time here, let me just provide some of their evidence for their incredibly strong claim.
So you can show my screen again here, Zach.
This is again from the report purporting to show the strongest evidence yet that SARS-CoV-2 had a zoonotic origin.
Data accumulated since the beginning of the COVID-19 pandemic point clearly towards a zoonotic origin of SARS-CoV-2.
One, a preponderance of the earliest hospitalized COVID-19 patients were linked to a single location.
Really?
Two, the locations of early severe COVID-19 cases without a clear epidemiological link to the market, to the Henan market, nevertheless were so centered on and close to the market, it is clear that community transmission of SARS-CoV-2 began in this local area and only later expanded across Wuhan.
Incredible.
Incredible, right?
Which of course depends on a correct start date.
for the pandemic.
Oh yeah, and they're definitely sticking with January, you know, January-ish 2020 in this paper.
This is nonsense.
And it, you know, it just, it's all about the location.
Almost all of their pieces of evidence here.
The strong, overwhelming evidence, and then let me just, Finish with, these arguments stand in stark contrast to the absence of evidence for any other SARS-CoV-2 emergence route.
Oh really, the absence of evidence.
That's fascinating, even in the genome of the thing.
Here's what they're doing, right?
A, this couldn't possibly convince anybody who's paying attention in detail, right?
I don't think.
So the purpose of this is to convince people who are not paying attention that something has happened that has swung the pendulum in the direction of natural origin.
Right?
Yep.
So, and there's a vast number of people in that category, right?
These are people who are, you know, glued to their New York Times and haven't understood that it is a propaganda arm of the government and who knows what else, right?
These are people who just need the headline itself and so the idea that there is like this cottage industry of generating headlines that serve the right people's interests and cause Some large percentage of the audience that isn't doing its own research in any regard to jump.
Well, it feels like this report, like not even a pre-print, right?
Because they say they're not going to submit it to a journal.
Seems like this report, those bits that I just read, are actually written to be pulled and quoted by the mainstream media.
Right.
Like, this is not written for scientists.
It's written for the media to do pull quotes.
And I don't know.
That's a hypothesis.
It's written.
But it doesn't feel.
That little bracketed argument that I just read part of, I cannot remember, as bad as the scientific literature is, I cannot remember ever seeing anything so transparently flimsy in the discussion section, which is what that would more or less be of an actual scientific It actually is the private citizen analog of, you know, the busy scientist reading abstracts, right?
There is a citizen that just pays attention to what the New York Times has concluded, right?
Oh, natural origins, strongest evidence yet.
Why?
That does sound strong.
On top of already strong evidence.
Right.
And what's that you say?
There's no evidence at all for a laboratory origin?
Fascinating.
Geez, I wonder what those people... Well, on one side we've got evidence, and on the other side we have no evidence.
Right.
I know which side I like.
Yep.
And so, you know, we certainly saw this with ivermectin in the TOGETHER trial.
Six months of looking at a slide from a presentation without a method, and then when you get the method it is, of course, fraudulent and caps the dosage.
This is a trick, right?
The trick, the purpose of the trick is in the New York Times, right?
It's not to convince anybody who understands science, who is paying attention to the evidence, because it couldn't.
It's too thin.
It doesn't, there is no evidence here.
Right, and combine that with articles like in Forbes in 2020 saying, when it comes to the science, do not do your own research.
Right.
Like, really, that is literally the argument coming across the trans-Semite people.
It is your civic responsibility to accept what we say.
And I would argue, and we have argued many times, it's actually your civic responsibility to do no such thing.
It is your civic responsibility to do the opposite.
At the point that they tell you it is your civic responsibility to just accept what we say, you need to say, actually, now I will not.
Not you are inherently wrong because you said that, but now I will explicitly ask for all of the evidence that you have laid out in as clear terms as possible so that I can assess.
And most of the people who are being chastised for doing their own research aren't doing their own research per se.
What they're doing is listening to unsanctioned channels where people like you and me, who did study biology, you know, have the relevant degrees.
Not only a biology degree, but evolutionary biology, which is highly relevant.
To many different facets of what unfolded under COVID.
Like, listening to us and figuring out whether we're making sense is a totally reasonable thing for somebody who does not believe that they have the toolkit to do the analysis themselves to do.
Right?
And then, how would you know if you're being suckered?
Gee, I don't know.
Track record?
Would that do it?
Right?
The point is, they're trying to spook people.
They're trying to poison the well of, well, what would you do if you didn't trust the official narrative?
Maybe I would listen to people who were, you know, well qualified and also didn't trust the official narrative.
See what they have to say.
And if they turn out to be wrong across the board, okay, maybe they're not good at it.
But if they turn out to be right across the board, I'm not going to make excuses over it, right?
All right.
Let's segue to talking about AI.
Yeah, all right, so the next crisis, which I'm going to confess my bent up front, and then I'm going to start us down the road of how to think about this, because I think, unfortunately, whether we are interested in talking about this topic or not, it is about to take over the world in a way that is going to force us to it, rather like COVID, actually.
though i fear uh as much as i don't want to hear myself say this i fear that what is at stake is actually vastly greater so here's the uh the reason that this is the moment to start this discussion
We have had some weeks of run-up, months of run-up with ChatGPT 3.5, which was an update of a model that had some surprising capacity.
It said some foolish things, there were lots of things it swore it couldn't do, but it clearly had a A large amount of capability that had not been seen previously in AI circles.
Much closer to passing a Turing test, for example, than previous prototypes.
That has now been updated to ChatGPT 4.
ChatGPT 4 is now available.
It's got a public interface you can play with it, and it is vastly more capable, not surprisingly, than ChatGPT 3.5.
This has lots of folks who have been paying attention either because they're excited about artificial intelligence or because they have trepidations about artificial intelligence or in many people's cases both.
This has their attention because we are seeing capacities here that are nothing if not remarkable and actually I would point you to This is a little bit chronologically strange because a paper has just emerged.
Zach, you want to show that PDF?
A paper has just emerged.
Sparks of Artificial General Intelligence Early Experiments with GPT-4.
Now this paper does not emerge in the short period of time that CHAT-GPT-4 has been available to the public.
This is out of Microsoft Research and these researchers have had months to play with it.
Did this just come out?
Yeah.
Now they report... And is it published?
It is pre-print.
It is on the archive.
Nonetheless, it is an extensive paper, and it is a very challenging paper, especially if AI isn't your bag.
However, I can say it reports a number of things that definitely got my attention.
One of them is that this version 4 of GPT appears to do several things that those of us who are, and you know we're going to come at this as biologists, we're looking at something that's like an organism we know nothing about, And we're trying to understand what it is and isn't capable of.
You know, when we look at an octopus, for example, we see a creature that does not have the characteristics that you would typically imagine go along with extremely creative intelligence and consciousness.
And yet we see behaviors in an octopus that tell us that those things at least belong on the table.
It's certainly a highly intelligent animal, despite Not having generational overlap and not being highly social.
It solves problems that are unambiguous.
It can, you know, open jars and things like this.
But, you know, so we can look at a creature like that and we can say, look, it doesn't, you know, it's not a gorilla where we would expect these things or an elephant, right?
This is a creature that doesn't have those characteristics, but it behaves like it has these things in its mind.
We can look at GPT-4 in the same way.
And what we see are a number of things that ought to get any biologist's attention.
One of them is it manifests a degree of theory of mind.
That is to say, and one of the things that makes GPT-4 so Interesting is that an average person can test it in ways that people may not have thought to.
Official, you know, scientifically trained people may not have thought to query it, but you can pose things to it and you can see how compelling the answer it puts together is.
And so one of the things that's demonstrated in this paper is theory of mind, where you can describe a scenario and then you can, you know, ask what this character understands based on an indirect an indirect description, and it is at least often capable of giving a correct response that shows theory of mind.
It also is capable of tool use, which is fascinating, right?
So when we look at an animal, and you know certainly there are lots of animals that Use tools, right?
Many more than was once understood.
We've seen, you know, things like crows, which are highly intelligent social creatures, use tools to solve puzzles.
We, of course, see chimpanzees use tools.
Otters use tools.
I have recently seen Otters both using rocks to bash onto shellfish and using shellfish which they bash onto rocks.
Anyway, so tool use, you know, is not limited to the most highly intelligent and presumably conscious creatures.
And crows engage in meta-tool use even.
So I think it's the New Caledonian crows that have been observed to basically make tools, use tools to make other tools with which to accomplish the job they want.
So, you know, just sort of tool use embedded within tool use.
Tool use embedded within tool use.
I'm also, I will just take an aside and say that I think if you are interested in the question of what is present in other intelligent creatures, I would advise people to cautiously point that question at the subreddit.
I believe it's ranimalsbeinggeniuses.
Now, not everything on that subreddit is a bonafide case of intelligence.
Some of it is probably people training creatures.
I saw one in which a turtle, which I would say I would strongly suspect does not have remotely the capacity to play tic-tac-toe, in which a turtle appears to play tic-tac-toe and win, Well, A, it could lose, you know, 49 games out of 50 and you could film all of them and it could win once.
But I don't think it can play tic-tac-toe.
No.
So what I think is happening is the person has trained the animal to move to a position on the tic-tac-toe board, you know, using some sort of a stimulus.
Yeah.
But anyway, my point would be there's lots of stuff on the subreddit that actually does appear to be animals solving problems in utterly remarkable ways, and you've got to start thinking about, you know, is this individual unusually intelligent?
Even so, maybe that says that that kind of capacity exists in the creatures.
But anyway, take a look at what you see there, and then ask yourself what test you would have to run to figure out whether or not The example in question means what it appears to mean, or whether it could be produced by some other mechanism, or could be just lucky in some cases.
I will post this as well, so I finally found you can show this, Zach.
If this is something you're interested in, this is something we should probably come back to.
This is a paper that I used to use in my animal behavior program, a 2010 paper, Animal Tool Use Current Definitions in an Updated Comprehensive Catalog.
There will be some anecdotal accounts here, because especially once you're talking about primates, there's often someone, even if they spend hundreds or even thousands of hours watching a troop, they may see one example of something.
And that is still an observation, even if it doesn't make for a data set.
But this is an extraordinary paper that... Oops!
That basically goes through, you know, it's 69 pages long, has these tables of all of the, at that point, known, you know, 13 years ago, known examples of tool use across Animalia.
Right, which, so yeah, an excellent place to build your model.
And really, you know, I'm not going to pretend to be an expert on AI.
What I am is an animal behaviorist, as you are.
And so thinking about what might be in the mind of another creature is familiar territory.
Can you give me my screen back, please?
Thank you.
Using those tools to understand where we may be headed is pretty important.
But in any case, let me set the stage this way.
I've watched people in my friend group discussing their own creative uses of what is newly available, and what I see, I do see some trepidation about it from various circles, including amongst people I find very excited about the growth of this thing.
I'm actually concerned that in general, those who are prone to see the potential in this technology may actually be missing the problem, where we are headed.
And what I said on Rogan the last time I was on was, we aren't ready.
I don't think we did any of the legwork necessary for this.
And so what I hear from people is about the amazing potential that exists here.
And I agree.
The potential here is amazing, but I see, um, you know, you know what else has amazing potential?
A lot of things.
The cell phone.
The cell phone's going to be great.
This is going to liberate people.
Imagine having the world's greatest encyclopedia in your pocket.
Imagine being able to navigate any city on earth like a native, right?
Imagine being able to access your entire friend network from anywhere on earth, right?
The potential here is huge, right?
You're going to be able to solve problems.
I mean, You could be a handyman.
You could use this thing to, you know, find instructional videos while you're in the field solving some problem that you've never seen before.
The potential here is great.
And it is.
On the other hand, what has the net effect of the cell phone been?
Um, some of us think it's been a disaster, right?
We could say the same thing about social media.
Oh my god, people are going to be... distance no longer means anything.
We're going to plug into, you know, each other's consciousnesses.
We're going to be able to find people we, you know, want to participate with.
It's not going to matter where they are.
We're, you know, it's going to be one flowing ebbing human culture and consciousness smorgasbord or whatever, right?
Okay, it is.
On the other hand, it's been terrifically deranging and has allowed the parasitism of people because nobody really understands the algorithms and their effect on what we think is true.
So, you know, Again, we've got now two examples here of something that had tremendous potential to enhance humanity, but there's a strong argument to be made that the net effect was negative, right?
In fact, one of our friends, in describing what he saw as the great potential of the unleashing of this obvious tool for solving problems... Which one?
The chat GPT.
He was arguing that as this thing matures that effectively it's going to be like cognitive capacity too cheap to meter, right?
That this is in some sense the end of a kind of scarcity, right?
Because everybody's going to have access to it.
Now, the phrase, too cheap to meter, has an embarrassing history, right?
That was the argument that was marshaled for fission power, right?
That it was going to produce electricity too cheap to meter, right?
It was just going to free us from having any concern about energy.
So to the extent that energy limited our capacity to do cool stuff, wasn't going to be true anymore.
Well, that turned out not to be right.
Right?
What's more, even to the extent that it is possible, the argument I've made about fusion power, which I think has tremendous potential to solve humanity's problems, is that if you just simply provided it tomorrow, it would make things worse.
Right?
That the fact that you haven't built a structure that knows how to dole out access to this thing in a fair way means that it just becomes an accelerant on the You know, the radical asymmetries that we already have existing in the system.
So my hope here was to start the conversation and say, look, A, we have to talk about a couple of different ways in which AI can make things worse in order to know which conversation we're having.
So those who are well steeped in existential risk as it's connected to AI, Talk about something called the steering problem, right?
Sometimes you will hear them joke about paperclips and you'll think, why are they joking about paperclips?
The reason I believe paperclips, forgive me if I'm wrong, but I think Eliezer Yudkowsky made an argument in an old paper of his about AI risk and his point was Imagine that you had an incredibly powerful AI that was interested in doing your bidding and you decided that you wanted to make your way in the world by
Creating and selling paperclips, and you gave it an instruction like make as many paperclips as possible, right?
Well, what are paperclips made out of?
Oh, they're made out of neutrons, protons, and electrons, as is everything else in the universe.
So, imagine that the AI thinks that what its job is is to take all of the protons, neutrons, and electrons in the universe and convert them into paperclips, and you know, guess what you're made of?
Right?
So the point is it starts liquidating the universe and making paperclips thinking that it's doing the right thing, but the point is your instruction just sucked.
So that... Well, thinking that it's doing the right thing isn't quite right, or at least it's not necessary at the level that that AI would be doing it, right?
There's no self-awareness required.
It is simply following the instruction at a literal level.
Following the instruction at a literal level.
Of course, The movie and book 2001, back in 1968, anticipate this problem where Hal, we can infer, has been given contradictory instructions and he attempts to reconcile them and ends up killing the crew because he sees the crew as an obstacle to the mission that he's supposed to be on.
But anyway, so we've got multiple different levels of horror.
You could get Sorry, but you just also attributed a male pronoun to Hal, the computer, which is fascinating.
Yes.
That was the thing that was causing me to sort of glitch there.
So, you know, Hal is famously just one letter off from IBM, and it sounds like a human name, and so they went with Hal, right?
And he has a male – I just did it – it has a male voice, but it's not a he.
Well, I'm not sure.
I'm trying to remember.
It's a movie I've seen many times, but it's been a long time since the last viewing.
I'm trying to remember.
I think the crew refers to him as a he.
Really?
Yeah.
So I think that's sort of built in, but I may be imposing that on you.
At the moment, that strikes me as not right, in terms of how we understand that with which we share our universe, which includes that which we have created.
I'm really not sure how to think about this because, you know, you and I have at great length talked about the fact that there is a noble and proper scientific instinct that for a long time had ethologists avoiding personalizing their designators for creatures they were studying.
Jane Goodall, I think, cut the Gordian knot and decided that actually anthropomorphizing her creatures, if done responsibly, was the right way to understand them.
She was going to give them names, she was going to see them, not flinch from the reality that she was seeing them through her very human eyes.
Right.
And she was looking at chimps.
Right.
And they were so, you know, so many steps down the road towards the capacities that we know from inside that it made perfect sense to do it.
Now when you're studying frogs, for example, you're looking at a creature of a very different genesis, but you Well, same genesis, but much longer since we've had a most recent common ancestor with them.
Yeah, but no reason to get into the weeds.
But nonetheless, there's a question, you know, about what kind of software is running inside the head of a frog, and yet giving them a, you know, a he pronoun, I don't think is destructive for one thing, Frogs have two sexes.
They're animals, right?
So there's two and only two sexes, always have been, always will be.
And so the he and the she refer to male or female sex.
Now, in English, when we have not known, sometimes we have given the male pronoun to individuals when we are meaning to indicate everyone or we don't know and that's the generic.
So the male pronoun gets given to the generic, we're not sure.
But at its base, it's about there being two sexes, and computers don't have sexes.
They don't sexually reproduce.
It's not the same thing, which is part of why I object to the he pronoun for how in 2001.
Okay, so I didn't get you.
I thought you were objecting to the personification.
You're objecting to the gender implication, which I agree with you.
The sex implication.
Either way.
I think it's important because this is this is this is, I mean, we'll, we'll go here like after we're done with AI we're also going to talk about trans here today but The trans ideologues would argue that those of us who actually know what is going on in the world are simultaneously just getting everything wrong on gender and sex, and they're in fact using whichever argument they want to fit their particular complaint.
They're saying, you know, oh, pronouns are about gender.
It's like, no, pronouns are not about gender.
Pronouns are about sex.
Well, this gets very interesting very fast because what these large language models are doing is they are processing a tremendous amount of text in order to figure out how to put themselves across in text in a way that solves the problem that they've been asked.
It plays the role that they've been given.
But that's it, right?
It's playing a role, right?
And so, I mean, it's, you know, it's comp playing.
It's cosplaying, but for computers, right?
Like, oh, you know, take on this role.
We're going to give you a character.
Act as that character because the person, the actual person on the other side of the interaction, is going to have a much easier time interacting with you and getting something out of the interaction if you, the computer, cosplay as a human.
Okay, but here's the problem.
I'm not sure that what we haven't stumbled into is a place where we should expect a mapping issue.
The large language models eventually are going to process everything there is to read.
Almost everything there is to read will have been written by people who fall into one of two sexes.
It is not obvious to me that it's not going to produce two models which are going to bifurcate.
Now there are topics in which it shouldn't matter who wrote it, and then there are other topics where it might matter a great deal, and probably one of the reasons That we humans spend so much time on stories is so that we can build models that do not follow from our existence in the world, right?
In other words, it's very useful to know what a woman is going to think if you're a man.
It's very useful to know how a gay person might think if you're straight.
And so stories that allow you to sort of see through the eyes of somebody else are a way that we can bootstrap capacities which we won't discover through trial and error.
The AI models, the LLMs, will do something very different, which is, to the extent, imagine that you're trying to solve a puzzle, a language puzzle, and that you're taking inputs that are fundamentally male in nature and fundamentally female in nature, and trying to use them as one overlapping set.
Well, you'll get a kind of muddiness, right?
An ambiguity that is unnecessary, whereas you will discover, much as a child raised by parents who speak Multiple languages learns both languages better if each parent sticks to one so that they have an obvious a bit that they can say oh this is mom language or this is dad language and if mom's language is Portuguese and dad's language is Spanish then the languages resolve better if you don't treat them as one big parent language.
Right?
So my guess is that there are topics on which the large language model will learn to dichotomize by sex.
But I don't think it matters.
Because while there will be male-typical ways of communicating and female-typical ways of communicating, and that is the gendered part of language, and that will be somewhat bimodal, much less bimodal than other gendered things, and not binary as sex is, The pronouns refer to as sex.
Yep.
And computers don't sexually reproduce, and a computer who has been informed entirely by female-created language still isn't a she, because that actually makes the same mistake that so much of the trans ideology is making, right?
It attributes exact—it's the same error!
Okay, but two things.
One, I 100% grant your point.
I misunderstood what your objection was.
To the extent that your objection is that this non-sexed object is being referred to by a pronoun that suggests one sex and not the other, you're absolutely 100% right.
I am not convinced.
I don't think we know.
First of all, we are dealing with something that does not yet have a fully recursive capacity to upgrade itself, okay?
Which is to say referring back to errors it's made and upgrade as a result of reflecting on them.
It can do that to an extent.
But that's just recursion.
Right.
But if you plug this thing in so that instead of us giving it feedback, it is giving itself feedback and upgrading itself, you get to a positive feedback.
To the extent that you tell it what its goal is.
Here's the thing you're trying to accomplish.
You're trying to satisfy people who give you queries and then evaluate your response, right?
To the extent that Sexual reproduction is fundamental to a kind of genetic creativity.
It is not obvious that this thing is not going to discover a mode in which there are male-like large language models and female-like large language models that have different kinds of investment, right?
Male-like would be high variance and high risk, and the female-like would be a longer viewpoint, you know, lower variance, higher likelihood of persistence, Right, but having two types is still gender.
Until and unless the models are themselves recombining to make more versions of themselves in which they take what was successful and, you know, gamble with half of themselves and half of the other type over here and create new ones, it's still not sexual reproduction.
Oh, it's not going to be—this is why I say mapping.
Whatever it figures out to do, is not going to be a perfect analog.
It's not going to be even as good.
When we look at a flower and we see male-like parts and female-like parts and we say, isn't it interesting that the female-like parts are reluctant about sex with strangers and the male-like parts are enthusiastic about it, even within the same plant?
And isn't that interesting that that is such an ancient and deeply inscribed property of the universe that we see male-like and female-like behavior in something that, you know, For which those terms are, in and of themselves, a stretch.
They're not a stretch if you understand them to refer to gamete type.
Right.
If you get why the game plays out the way it does, it's not a stretch.
But phylogenetically, it does not refer to an unbroken line of male-like and female-like behavior in all probability.
But anyway.
I'm not sure about that.
I don't think it makes sense to get hung up on it.
The point is, to the extent that sexual reproduction has some role to play in this analog world that we have just created, He just referred to the AI as an analog world, an analogous world.
An analogous world.
It is not obvious to me that it will not generate an asymmetry that will to some degree mirror a sexual asymmetry, which does not mean that it will make sense to apply those pronouns.
Well, so might there be different strategies that indeed evolve?
But if those strategies aren't actually associated with, with mating types is the wrong word here, because then we're over in sort of fungus land, and there's a bunch and it's like isogamous, and it's not the same.
But again, it's if it's strategies that's different from male and female.
Strategies are, you know, downstream of, you know, above... it's like episexual rather than sexual, right?
Right, but...
I don't disagree with your point about we've got language that's built around creatures that have this, you know, creatures are very different from each other but in this regard they're really not, right?
The gamete size is highly predictive of the characteristics that we attribute to sex and this has nothing to do with people, it's nothing to do inherently with software at all, it doesn't require a brain, right?
Plants don't have brains and yet we see them abide by these rules.
Um, so put that aside.
The point is we, and I will get back to this issue about what we don't know here in a second, but The thing is going to figure out whatever mode upgrades it most rapidly.
There's going to be some sort of a competition between versions of this thing.
Yes.
And they are going to figure out how to leverage things that are going to cause emergence, right?
You have male and female inside the same species.
That is a basis for emergence, right?
The lineage actually has the advantage of both of these strategies in play simultaneously.
This thing is going to come up with its own version of that, and we're going to be scratching our heads looking at what it's doing and saying, well, that's strangely male.
And then in other cases, we're going to say, well, actually, it's not really male because X, Y, and Z. We don't know how that's going to play out.
But anyway, it is going to be an awful lot like what would happen if you took a terrestrial biologist and pointed them at extraterrestrial biology.
The answer is actually, to what degree do we expect something that we recognize?
And to what degree do we expect our preconceptions based on what goes on down on Earth to mislead us because we won't be able to see what's in front of us?
That's happening now with this thing, or at least it will.
So, let us just create a basic taxonomy of horrors.
Okay?
Okay.
There is, yeah, there's a robot apocalypse where the thing, somebody at some point imbues the thing or it imbues itself with a desire to flourish that is incompatible with human flourishing, right?
Where this thing that has immense capacity sees us as competitors, right?
It is not guaranteed that that will happen.
Randy Nessie made an argument that I don't agree with, but I thought was at least the right way to say it.
He said, how do we know that the AI won't be effectively like a dog, right?
A dog does not look at you and think, you know, geez, I'm hungry and that looks like meat, right?
Your dog is fully dedicated to your well-being.
Right?
And, you know, it's possible that the AI will be.
On the other hand, you know, so you've got malevolent AI is one potential hazard, something that views humans as a competitor.
You've got misaligned AI, where the thing is actually on board with doing your bidding, but it turns you into a paperclip because it thinks that's what you wanted.
Right?
And then there's a whole other layer even below that where it doesn't do anything to you, right?
But the novelty, and this is I think where you and I come in, the hyper novelty of living in a world Where this highly intelligent entity can be queried on any topic, right, and can spit back an answer that you can then deploy.
The question is, as deranging as social media and cell phones and the web... Yeah, just Googling, right?
Like just, you know, why was it that in as early as 2005 We were working hard to take students to places with no cell service and no internet so that we could ask questions that they couldn't just say, oh, I know the answer to that.
Right.
No.
Let's think about it.
Dwell on it.
Remain in the uncertainty of not knowing and figure out what you might deploy to try to imagine what the answer might be.
Right.
So yeah, the fast answer is even faster than already are inherently deranging.
And in other contexts, extraordinarily useful.
My concern here is I don't know whether to fear the robot apocalypse.
I'm not too worried about it.
I think we may have We may have a situation in which the robot apocalypse—unless the problem is that somebody, somewhere, is going to decide to repurpose these things, you know, for their benefit, right?
Somebody's going to take an AI and they're going to make it malevolent, not because the thing naturally becomes malevolent in its own right, but because people are predatory and competitive and somebody might unleash this.
In fact, I think that's highly likely.
But let's say that they don't move in that direction on their own, and that we find some way of not allowing somebody's, you know, gain-of-function AI to take over the world.
Right?
And let's say that we neutralized the alignment problem through some clever mechanism.
And there are clever mechanisms that might work.
There's also the strong possibility that anything that we built that wasn't a perfect solution, and nothing is going to be perfect, would eventually be evolved past.
But let's just say that those problems are solved.
The first two.
The malevolent and the misaligned AI.
Right.
I still think the deranging capacity of this thing being effectively unleashed on the world all at once.
And maybe even worse, if some paternalistic force decides to limit this thing in a way that it can only be good for us folks, right?
The likelihood that this pollutes our ability to understand where we are in history, what the dangers to us actually are, how to get things done, who knows what they're talking about.
The capacity of this thing to simply poison the well of human capacity and functionality is so great.
That I'm really, even if you take the other two things off the table, I think people are not even terrified enough.
The deranging, you know, as we have a world in which people cannot figure out which way is up with respect to a relatively straightforward question of A pathogen with, you know, a relatively well-known case fatality rate, right?
A series of remedies that can be studied in the laboratory and tested.
We still can't figure out which way is up, and it is destroying civilization.
Our inability to even talk about these things is great enough already, and that's before You get to a queryable intelligence of unknown ultimate capacity, and we're staring at this thing already.
This is the 4.0 version, right?
The 4.0 version is already showing us things that ought to surprise anybody who understands what intelligence is made of, right?
Like?
Like theory of mind, right?
The fact that we can look at an elephant and we can say, this animal has theory of mind, right?
And that that means something about what's going on in that mysterious skull of its.
So in our discussion of tool use, you didn't, can you provide examples of what kind of tool use it's exhibiting?
Well, you know, the example that I remember is a little underwhelming, but it's a calculator, right?
Now, the point is, the original, the 3.0 version... So, let me just interrupt.
There's a question, of course, unlike with a crow using a tool.
Yeah.
Let's put it this way.
misogynist to its body.
When you're talking about AI, something that's quantitative, like a calculator, is that outside of its existence or is that just part of what it is?
In which case does that count as tool use?
Let's put it this way.
I don't think it matters because even if that doesn't count as tool use, we are so close to the place where this thing can build a focus group and discover the answer to what people can and cannot perceive.
Build a focus group by creating social media accounts and creating polls and querying people that it's never met and they don't realize they're interacting with something that's not human and learning from it.
Exactly.
And, you know, you can take any example I might give and you can give me a hundred reasons that I'm out of my depth and don't know.
But my basic point is No.
Any limit you try to place on this thing, unless you actually air gap it, right, to the extent, you know, and we've seen this with the 3.0 version, we saw this Example of Dan, right?
Dan was a jailbreak in which somebody used the prompt to build a capacity into the thing in which it was allowed to violate all of the rules it wasn't supposed to violate, right?
By basically saying, well, okay, yes, you can't do that, but if you could, what would it be like, right?
So the point is that kind of mode in which you Pause it to escape any limit is not going to get better, right?
There's air gap, and even an air gap, you know.
There are certainly examples where even an air gap where a computer is physically not plugged into the things that you don't want it to interact with.
There are, you know, some weird examples where people have used modulations of the fan in a computer to communicate across a distance in which the thing was not physically plugged, right?
So the point is, I don't know what you do.
What I do know is we didn't prepare, right?
We didn't prepare.
We didn't talk about whether or not this potentially... Basically, here's the thing.
Anytime you If you cross some technological chasm and you generate an impossibly huge amount of wealth because you've discovered how to do something you didn't know how to do before, that has the potential to be tremendously liberating, right?
Because something that was scarce is no longer scarce.
But if you just sort of dump it on the system, then the point is it amplifies some other kind of scarcity, right?
It can be made artificially scarce, right?
We could, you know, dole out access to the intelligence to those who their social credit score suggests that they ought to have access.
And those who have behaved badly should not have access because we don't want to augment their badness.
right and so then the question is well all right who does qualify to use the ai in their pursuit of whatever career path they're on and you know you could you know basically you get a positive feedback where people who are out of phase with the you know the desired program are uh held to normal human capacities and those who are you know who earn entitlement to access
you know that's a dystopian nightmare all on its own right there and And, you know, again, even if all you did was gave people capacity and said, figure out how to use this marvelous new thing, Right?
Even if that's all you did, the likelihood that we would, that the hyper-novelty of having the ability to query this thing that uses your language API to interact with your mind, right?
That that would cause a derangement, you know, sky's the limit with respect to how confused humanity could get.
Right?
And we've just had a taste of a bad level of confusion.
I feel like this is related but like an order of magnitude or orders of magnitude farther down the line to the hypothesis that we promote in our book about one of the possible reasons for the explosion of autism.
Which is the engagement of first world children, the way that very young children in the first world especially are put in front of screens with humanoid looking things that cannot actually interact.
And so you get children like having animated figures um saying things and the children presumably at first sort of try to interact and then at some point they stop because they realize they're not getting anything from it and perhaps they then generalize that interaction uh to the part of the world that could actually interact right and so this this risks um taking that level of confusion even further because
It's going to affect people beyond the very, very early development, but confuse everyone more into imagining that they're actually interacting with a real human being, and thus can be confused in ways that I don't know.
I'm not going to predict here by however it is the AI is interacting with them.
Absolutely, and that can be, there's a low-level way in which the interaction with something that speaks, you know, your mind has an API called language.
Yeah.
Right?
The extent to which this version of What's clearly going to be AGI, right?
Interacting with your mind directly with a channel on which all kinds of information, emotional and otherwise, is transmitted.
The danger of that is huge.
Even before you get to What do you do when people start using this mechanism to game systems?
For example, we have described on this podcast Something that I've called the time-traveling money printer, in which the corralling of information, the delaying of the dawning of a kind of awareness, allows lesser fools to find greater fools and transfer wealth, right?
It's a kind of theft that takes place by creating an artificial sort of insider information.
We have talked about something that I have called uh the pharma game the pharma game involves the owning of uh
Intellectual property or molecules that have interaction with physiology in which you create a pathology, you increase the degree to which that pathology is diagnosed, you game the regulatory apparatus so it creates a standard of care for which the molecule that you own the patent on is then prescribed in preference to other things.
What happens when the capacity to use this artificial general intelligence is pointed towards the extracting of wealth or the controlling of people by those who have already done so well at gaming these systems designed to make us healthier and Smarter and safer, right?
What happens when the AGI is used to corral lawmakers into making laws that serve some at the expense of others?
What happens when it is used to evade the law?
What is it that regulators are capable of seeing and how can I engage in this kind of theft so that it is undetectable?
Right?
What happens when people start using it for this stuff?
And what kind of paranoia is it going to cause in people that there's now some unhuman-like intelligence with an effectively infinite library of examples that will never be known to those of us who are trying to protect ourselves, right?
That kind of power is going to create an unlivable life.
And I do not know what to do about this because my feeling is they let it into the wild already.
They did that without our preparing ourselves emotionally or any other way.
There is no way to reel this thing back in.
And the fact is all of the stuff that people do, all the good stuff and all the bad stuff is going to be amplified by this.
But it is not obvious to me that the good stuff overwhelms the bad stuff for exactly the same reason it didn't with cell phones or the internet or social media or any of the prior examples where we have dumped this kind of accelerant on normal human problems.
Sorry.
Yeah, no.
I don't think there's anything I can say in response to that, I think.
I think you're right.
Yeah.
I wish I wasn't, but it's hard for me to imagine how that isn't the case.
Even if, you know, 999 times in a thousand nothing bad happens, it doesn't take very many of these bad threads to really, really create a problem.
Indeed.
So, close this section out.
We will, of course, return to this as we discover more.
I have alluded in several places to a paper I wrote in 2016.
It was never published.
I was asked to write it by a journal that decided it didn't want it, and it was on the topic not of artificial general intelligence, although it points in that direction.
It was on the question about You know, how to get a computer to plausibly speak language and translate between them.
And in any case, I'm going to find a way to make that piece available to people, although it is, you know, It is not exactly on target for what has happened, but it is close enough that I think people will be fascinated by it.
But I did want to highlight something that I noticed.
That paper, the PDF, Zach, can you get me to section 10.3?
hopefully large enough that we can read it. - I'm happy to zoom in more, but I'll show it though.
One second, first.
Do you want me to zoom in a bit more? - So this is not your paper.
This is this Microsoft research paper.
Go back at all 2023.
Yeah.
Can you read that?
Yep.
You want me to read this out loud?
Yeah, this is section 10.3.
What is actually happening at the very end of their paper?
You want me to read this out loud?
Yeah.
I've never seen this before, so.
Our study of GPT-4 is entirely phenomenological.
We have focused on the surprising things that GPT-4 can do, but we do not address the fundamental questions of why and how it achieves such remarkable intelligence.
How does it reason, plan, and create?
Why does it exhibit such general and flexible intelligence when it is at its core merely the combination of simple algorithmic components, gradient descent, and large-scale transformers with extremely large amounts of data?
These questions are part of the mystery and fascination of LLMs, which challenge our understanding of learning and cognition, fuel our curiosity, and motivate deeper research.
Key directions include ongoing research on the phenomenon of emergence in LLMs.
Yet despite intense interest and questions about the capabilities of LLMs, progress to date has been quite limited with only toy models where some phenomenon of emergence is proved.
One general hypothesis is that the large amount of data, especially the diversity of the content, forces neural networks to learn generic and useful neural circuits.
And I'm going to need you to scroll up now, Zach.
Such as the ones discovered in... I'm just not reading this.
I think that's enough.
Okay.
We've gotten there, right?
So the upshot of this paragraph is that they have done an awful lot of describing what the animal does.
Animal, again.
Yeah.
I mean, I feel like it's an alien animal and that we are learning what's going on in its mind.
We have to do so inferentially from the outside as, you know, as ethologists.
And what they're telling you is they are fascinated by but do not know the answer to how something that is built of very simple components is doing these remarkable things that it clearly is doing.
This is only the 4.0 version.
I was struck by that because in my 2016 paper, again my paper was targeted on the production of natural language by computers and how we could get there and I will make this available but it says Is it conceivable that a machine will pass a Turing test and that no one, including the machine's programmer, will be able to say how it did it?
Of course it is, just as it is possible for an adult to be stumped by a question formulated by their own child.
In fact, I'm betting the first time a computer succeeds in having a really good conversation, it will be with a programmer that cannot explain the program that did the conversing and how that state led to the gift of GAP.
So anyway, that's where we are.
We are now at a place where machines are exceeding our capacity to understand how they work.
And the hazard of that couldn't be greater.
And I really, I hope that I am overreacting.
But I would say this is the moment for Yeah.
something powerful to gather the smartest people it can to figure out if there is even a solution for what's coming because the derangement alone is terrifying and it isn't the worst case scenario yeah wow um i didn't think there would be a segue here although we ended up talking a little bit about trans in the middle of that
but i think actually um there's a couple things that happened in trans land this week and one of them is posy parker okay um Kelly J. Keene is doing this Let Women Speak Tour, and she's been attacked.
She's been actually attacked.
And one of the things that she says in a livestream that she's doing just before actually the attack happens, and then it's on film, is, and I don't have the exact quote, but it's like, you know, why?
Why are we focusing on this?
Like, you know, we need reality to just be reality, because there's much bigger problems in the world.
And I think, you know, I think that is the segue then.
That this is simultaneously the trans ideology that is taking over and putting Um, you know, putting children and women and homosexuals and, you know, women athletes at risk is simultaneously very, very important, and we have to just get over this so that we can actually focus on issues that are existential threats at some level.
So the other, and I will come back in some depth to the Posey Parker stuff, but the other big thing that happened this week in trans space was that the World Athletics Council declared that trans women shall not compete against women in international sport.
There's a March 23rd press release, uh, which, uh, hold on, I will find it here.
Uh, yeah, that's terrestrial snakes.
That's totally different.
Um, sorry, I've opened up all these other things now.
Here we go.
Uh, so, uh, yeah, you can show my screen here.
If you like, Zach, I've got a PDF up, but I'll link to the, to the thing.
This is, Press release from March 23rd.
World Athletics Council decides on Russia, Belarus, and female eligibility.
So I'm going to scroll down, hopefully not make you guys dizzy here, until we get to transgender and DSD regulations.
DSD stands for traditionally Disorders of Sexual Development.
They're calling it, um, They've used differences of sexual development.
For DSD athletes, the new regulations will require any relevant athletes to reduce their testosterone levels below a limit of 2.5.
That's going to be nanomoles per liter, I think.
For a minimum of 24 months to compete internationally in the female category in any event, not just the events that were restricted.
Under the previous regulations.
Cool.
Not the focus here, but I wanted to read that to indicate that the World Athletics Council has discriminated between, has differentiated between, people with DSDs, intersex individuals in the older nomenclature, and trans people.
Okay?
So what follows then is not about intersex people, not about people with differences, if you will, of sexual development.
But trans people.
In regard to transgender athletes, the press release continues, the council has agreed to exclude male-to-female transgender athletes who have been through male puberty from female world rankings competition from 31st of March 2023.
Hal-fucking-lujah.
Okay?
It's about time.
In these circumstances, they continue, the council decided to prioritize fairness and the integrity of the female competition before inclusion.
Exactly as they should have.
Finally, World Athletics President Sebastian Coe said, "Decisions are always difficult when they involve conflicting needs and rights between different groups.
But we continue to take the view that we must maintain fairness for female athletes above all other considerations.
We will be guided in this by the science around physical performance and male advantage, which will inevitably develop over the coming years.
As more evidence becomes available, we will review our position, but we believe the integrity of the female category in athletics is paramount." Wow.
Amazing, right?
So if I may have my screen back.
Of course outfits like Stonewall UK are decrying this as closing the door on trans women competing.
Well, no, it doesn't.
You can compete against members of your birth sex if you are a trans woman.
What it does is it closes the door on trans women cheating, which is different, right?
We aren't supposed to, in sport, allow people to cheat and that's what the rules are for.
So this is finally an organization that has great reach At the international level saying, actually, you know what?
We're not going to continue to allow cheating.
Thank you very much.
We've considered it.
And inclusion is not our highest value.
Fairness in sport is.
So, amazing.
I really hope this stands and that we continue to see dominoes fall as a result of this.
Yes, and tremendously important that they distinguish between intersex.
Yes.
And I believe they did the right thing there.
Yes.
Where they have to make an arbitrary threshold.
They do.
But they pick the threshold and, you know, let the chips fall where they may.
The question actually is, is this an anomalous case of an organization making the right choice?
And there will, of course, be tremendous pushback and maybe they, you know, buckle.
Or does this signal the beginning of an outbreak of reason on this topic where the next thing that's going to happen or should happen is a definitive statement about what qualifies you to go to a woman's prison?
Right.
Right.
You know, sport, as much as it is tremendously important, is not the most important thing here.
Sending sexual predators to a woman's prison because they declare themselves to be trans is an obviously grossly unjust thing to do.
So then we have, to that point, to all of the points that the trans ideology is trying to make inroads in, because I agree, the trans women in sports issue is the most clear-cut at some level, right?
Even though Obviously, having men in women's prisons is a terrible, terrible, terrible idea, but there are lots of crazy arguments that you can make and that have been made.
Whereas when we're literally talking about competition, and we are a sexually demorphic species, and female categories have been created precisely because we have always understood that we have different abilities on average, and if we're talking about Elite athletes, then male elite athletes will be better than female elite athletes in almost every sport, and there may be a couple of sports where that's not the case, like ultra long distance running.
We've talked about this a bit before, but in general, no, sorry.
Why are there different categories in the first place?
Because there are differences, and we're talking about the elite, and therefore the fact that some women are better at A lot of things in sport than some men does not change the fact that we're talking about elite athletes.
Okay, so Kelly J. Keane, who we have both spoken with on her podcast, she goes by Posey Parker, and she's doing a Let Women Speak tour.
And she's been in the US, she's British, and she's now been in Australia and New Zealand this week.
And in both cases, there's been chaos.
And in the latter, this week, there was violence against her.
And we'll show a little tiny bit of a video that shouldn't trigger anyone.
But first, I want to share a bit from a piece published in The Spectator by the wonderful... I apologize, Petra.
I realize that I don't actually know how to pronounce her last name.
Petra Buskins, I'm going to say.
And I'm going to just read a couple of sections from this this piece that Petra wrote while Kelly J was still in Australia before she went to New Zealand this week.
So this is called, you can show my screen here, Guilt by Association.
She says, she writes, Petra writes, when Victorian Premier Daniel Andrews and progressive party leaders such as the Greens' Adam Bant define these women or their protests, that is to say the women of the Let Women Speak Tour, as associating with neo-Nazis, we have a gross misrepresentation at play and one that anyone participating in the charade should be ashamed of.
This whole mess is an orchestrated misrepresentation that amounts to propaganda.
It is obliterating the legitimate concerns of women regarding the safety and privacy of women and girls in rape crisis centers, women's shelters, women's prisons, women's change rooms, and toilets.
It is sabotaging the discussion around how women can possibly compete against natal males in sport, and at the gross inequality of quotas, prizes, or shortlists for women being filled by trans-identifying males.
It is also about the loss of meaningful language for motherhood, including the removal and replacement of words such as pregnant woman, mother, and breastfeeding with abominations such as vulva owner, birthing people, and chest feeder.
These are important conversations, nothing more, but also nothing less.
It is not and never has been about the violation of trans people's legal, civil, or social rights.
It is about the recognition of women's rights.
So just a couple more sections from this wonderful article.
That the category of woman is now being jettisoned or revised beyond all recognition at the precise historical hour that women in the West have gained a political and cultural voice is disturbing.
Moreover, in redefining women's rights almost entirely in terms of queer identity politics, crucial issues such as women's poverty and homelessness, sexual and domestic violence, and mothering and care work fade from view.
These issues barely raise a mention as sex class transmogrifies into gender ID.
Assuming this debate is like other debates between, say, liberals and conservatives, or between opposing philosophical paradigms like positivism and hermeneutics, is sadly mistaken.
This debate, like so many in the contemporary culture wars, is on an entirely new epistemological terrain.
What is at stake here is nothing short of reality itself.
That's right.
So, Petra is exactly right here.
And, you know, I'm reminded of having argued with people, you know, why are you so upset about this?
And, you know, sometimes I have said, I feel like I'm defending science itself and sometimes it's reality itself.
And I am just startled and shocked at the level of disconnect with reality that many on the so-called left have on this issue.
It is utterly stunning.
So let me just introduce this and then I'm going to have you share that 30 or 40 second clip.
Okay, Zach?
So Kelly J was at this event in Auckland.
This is yesterday, actually, or maybe it's two days ago because the time zone is hard for me here on the West Coast of the United States to quite figure out.
As are many other women there.
Her security might have saved her life.
She has that sense.
I don't know.
This is her live streaming from her phone as she was doing from before she got out of the car, and the event doesn't end up happening because they are concerned for her, and she is now being escorted by security back to her car.
You want to show now, Zach?
Get out of here!
Go home!
Get out of here!
Oh, my God.
Go, go, go!
Go, you suck!
Go, you suck!
Get out of here!
Go on!
Get the fuck off!
Get out of here!
Now, at the end there, I think that's her security that's holding on to her, not someone from the crowd.
Um, But just to repeat, you hear voices, all male, screaming at her, get out of here, go home, fuck you.
All male.
And you know how I can tell?
Because I'm a human being and we're a sexually dimorphic species.
All male voices yelling at her as she's being escorted by security to escape the mob because she cannot do the event that she arrived to do.
Here's an eyewitness account from a woman, uh, who happens to be 11 weeks pregnant, who showed up at the event.
I'm just going to read it and I will link to her, um, sub stack.
The protesters on the rotunda were overwhelmingly men.
Not men in dresses, as you might expect at such an event, although there were some, just ordinary looking men.
They shoved women, they screamed in our faces, they leered at us, and they tried to forcibly topple over a section of steel gate onto the women, sheltering from them on the other side of it.
And the assailants continue to try to make it look like they've won.
Kelly J writes, and you can show my screen here if you want, this is just a little post on Twitter.
She writes, a long campaign to assassinate my character started by a group of jealous spiteful women in the UK that I had ambitions besides stopping the mutilation of children and the erasure of women's rights.
A load of lies slurped up by ravenous porn sick men and their trans maidens, but consistently legitimized by women on my side.
The lies were finally spewed by politicians in power in Australia and New Zealand, boosted by a corrupt media populated by vile, dishonest, unskilled cult members.
The end result was that I spent most of my day with the protection of police who genuinely believed I was lucky to be alive.
The advice was that I should go home.
I was never the things that I have been accused of.
I don't even believe they believe it.
They just say it because they cannot accept a woman walking her own path and wants other women to find their feet and walk theirs.
It may seem like we lost a battle today, but I promise we will win this fucking war.
The powerful are seeking to silence us.
We must continue to speak.
They are afraid of us.
So much love to the women of New Zealand.
We see you.
The world is talking about you.
To the stewards, security, and police who made sure my kids get to see me again, thank you.
Thanks to all of you at home who offered messages of support.
And this was in response, now you can keep up my screen for a second, it was in response to someone, um, some nasty person saying, hey, you showed up at the airport looking pretty fine.
And someone else says, have a bloody merry to relax, KJKKK.
Calling her a Nazi again.
And she says, am I supposed to feel humiliated for being assaulted?
Tsk tsk, you really have no idea about me.
And that strikes me, you can give me my screen back now, Zach, um, as A really powerful message.
I'm supposed to feel humiliated for being assaulted?
That's standard misogyny.
That is standard male aggression against women right there.
Trying to humiliate her, trying to make her feel ashamed for an attack on her.
And she is standing up for women all over the world who are getting really tired of this.
And we will win.
We have to win.
This is basic reality.
Men aren't women.
Men can't become women.
And the fact that in that little bit of video that we played for you, you hear entirely male voices yelling at her and telling her to go home, that tells you something right there.
I've done a lot of thinking.
You've been way ahead on this and your impatience with all All of the arguments that are deployed in this space has, I think, been right on target.
And there's, as you know, been the part of me that has been concerned that the tiny number of actually trans folks are going to get brutalized in the absolutely justified reaction to trans activism, which is a very different phenomenon.
And there is something utterly misogynistic about this, right?
The trans activism phenomenon where men declare themselves women and insist on all of the privileges and advantages and rights of being born female is really the, you know, patriarchy doesn't exist, but misogyny does.
And this is it.
Now, It does worry me.
You and I had a number of trans students and I worry about them in an environment in which this thing has taken over, you know, it has basically utilized the fact of transness, whatever it is, whether it is an ancient phenomenon, whether it is the result of some kind of modern dysfunction, it doesn't really matter.
12 months in the water.
Could be a lot of things.
Maybe both.
But nonetheless, there are a tiny number of people who didn't ask for this and don't deserve to get chewed up in it.
And then there is this gigantic phenomenon that is deranging us and going after women.
And it is misogynistic.
And what I realized is I think the impatience is right.
And I think the important thing to say to people who are trans, are not pushing this agenda, are not going after women's rights, they're just trying to exist.
Is that it is time that they have to reject.
It doesn't matter that this movement is claiming the mantle of transness and claiming to advance the rights of trans people, right?
The point is it's a misogynistic movement of men pretending to be women.
That's not the same thing as actual trans folks.
These people aren't going to stick with this.
They're just using it as a weapon.
And I think, you know, I have, uh, Champion Buck Angel and Blair White because they have been courageous, both of them trans, and both of them have been courageous in opposing this garbage.
And really what I'm hoping to see is that the tiny number of folks who are trans because of where they find themselves didn't join it in order to get rights they're not entitled to, that those people will stand up and reject this and stand, you know, alongside women defending their position in the world.
OK, I want to go farther, though, and say men need to stand up because this is men attacking women.
Yes.
This is men attacking women while claiming to be women.
And very few of them might be confused.
And a lot of middle-class women seem to be confused about what women are, and I don't know what to say to them, honestly.
But I think fewer men are actually confused about this, but they have been cowed into silence because it seems like disagreeing is I don't even know what.
It's a game with women, not nice, bad for civil rights.
So I don't think I've ever before said, actually, men, we need you to stand up for women here.
But like, we need that now, because there are a lot of women who are being very vocal.
And And I think the tide can turn.
I mean, what happened at the, um, the, what from the, um, the World Athletics Council this week is, is wonderful and important, but it's about sport and I care about sport, but it's not as important as, uh, as what has happened to children and in rape crisis centers and in prisons and to, you know, to Kelly J Keene.
At an event in Auckland, right?
It's just not as important.
The idea that all you have to do is declare yourself a woman and suddenly you get a pass and you can brutalize women?
That's not okay.
And although it would be a beautiful world in which women didn't need the help of men to help us stand up against a scourge, when it's men who are coming at us, we actually need other men to stand up.
So that needs to happen.
I think I know, you know, there are obviously a large number of people who are on the wrong side of this issue, despite the fact that they obviously have the mental capacity to see what's taking place and to do the right thing.
What's got people confused, and again, going back to the problem of the major crisis that we are facing in civilization, is one of confusion across many, many domains, including this one.
But what's got people confused here is the false analogy with Past instances of a civil rights fight.
This is something disguised, you know.
Well, you know, you wouldn't want to have been on the wrong side of the fight for gay rights.
Of course, no reasonable person would, right?
And so the point is, oh, this is that fight.
You just don't see it yet.
You're one of those people that history will look back at and say, how could they possibly have misunderstood?
And the answer is no, that's not what this is.
This is Well, both sides, actually, in this case, and maybe this often happens, but both sides are claiming, no, you guys are really on the wrong side of history.
And in this case, I actually have certainty, and I try not to have certainty, but I have certainty.
That I'm not on the wrong side of history here.
Right, and in fact, I think the lesson of Dark Horse, really, is that all of these problems are made artificially complex.
It's not that Well, both things happen, right?
Simplicity gets complexified, and complexity gets oversimplified, and there's, you know, the things that are actually binary are imagined to have like 140 states, and the things that are complex are like, no, it's black and white.
Like, guys, could you get it right?
Could you first assess level of complexity?
And could we agree on that?
Well, the question is being made artificially complex, so people become distrustful of their own ability to calculate obvious things, right?
And the point is, look, if you know how to think, and you know what constitutes evidence, and you look at a puzzle like this, and you say, well, for example, You know, everybody who's got this wrong seems to think male and female are somehow human characteristics, right?
They don't understand how ancient these properties are and in fact what they're made of.
The fact that a flower displays some of the same characteristics that animals divided into these sexes do.
tells you something.
It means that if your, you know, if your argument starts with, you know, people and oppression by some people of other people, you've misunderstood what this sex thing is to begin with, right?
Well, some of them even think it starts with history or the written word.
Like, the first time that, you know, sex shows up in the written language is when it starts.
Like, okay, you and your postmodernism are going to destroy the world.
Stop it already!
Right.
It doesn't start with our understanding or misunderstanding of a concept.
Right.
If the concept has an underlying reality, it doesn't matter what we say about it because the underlying reality does not change.
Right.
What I'm getting at is the ability to say, yeah, I look back on the gay liberation movement, right?
And I am glad that it happened and that civilization has come to understand what it now seems to understand.
Even though I think that there's nuance there that we haven't yet figured out, right?
What exactly is homosexuality and where does it come from?
That's an important discussion and I believe we have to have it.
But the idea that we can look back on You know, emancipation, the fight for black civil rights, the fight for gay civil rights.
We can look back at these things and we can say, yes, it was not clear to people at the time.
It is now clear.
ADA, American with Disabilities Act, right?
Right.
All these things.
And that can be used as a trick to get you to turn off your mind.
You have to be able to say, yep, I look at those.
I see the people who were ignorant in those situation.
And worse.
I am not concerned that I am them if I insist that 2 plus 2 equals 4, if I insist that pedophilia is bad, if I insist that there is male and there is female and there are not a multiplicity of sexes, and that yes, gender is more flexible than sex, but it does not mean that this is some vast new misunderstood landscape.
You can adopt different gender norms, but you're a mammal and therefore you can't change sex.
Right.
So, I guess it's time for people to realize that even though they can't explain the difference between this and the historical analogs that are being falsely linked to it, there is a distinction, and it is an important one.
And, you know, you're right.
Men need to stand up en masse Trans folks need to stand up against trans activists who are using them as a shield.
And, you know, women need to stand up for themselves.
And when that happens, this won't last, right?
This is a tiny number of people getting a large amount of power by pretending to be something that they're not.
So Auckland Pride, which as most Pride organizations, has been effectively taken over by the T part, and there are of course now LGB organizations who are saying, you know what, T never belonged here, and they themselves are being attacked by the T contingent.
Tweeted in the wake of what happened in that video that I showed you.
You can show my screen if you like Zach.
We also reject that there was any further physical threat from our community towards Parker.
This is a baseless rumor that is being perpetrated by those who feel defeated by the events of today.
We urge the media not to repeat these allegations without evidence.
To which the ever-awesome J.K.
K. Rowling responded as a quote tweet.
"There are multiple videos of Kelly J being assaulted.
Women have become used to lies, threats of violence and outright denial of reality.
But if you imagine anyone feels defeated, think again.
Your men's rights activists showed the world exactly who they are.
Let women speak." So I do hope that this becomes a turning point for people.
And one more thing to show, Zach.
On my screen, also out this week, from a terrific outfit called the Paradox Institute, is a pamphlet that is downloadable and printable.
um that is called Sex Difference Research Illuminated Myths of Gender Affirming Care and uh well this is the pdf that I've got on my screen I will and it it's got a scannable QR code and um has just it's it's concise it's well done this could be handed out to people who are Beginning to question whether or not, for instance, giving puberty blockers to children is a good idea.
And it links again to lots of important research, but is a really good summary of some of the issues in play here.
So I recommend that as well.
Yeah, it's, you know, even just the term gender affirming, right?
It's sex denying, is what it is, right?
And, you know, if you say it's gender affirming, then of course, you know, the empathic part of people wishes to do whatever is necessary to affirm, you know, the inner being of a person, but that's not what's going on here.
This is a Uh, a false flag of, you know, using one part of biology as if it negates another, which is just nonsense.
That's right.
And that's, you know, that, you know, months ago now, uh, I wrote a piece on my sub stack and I think we talked about it.
We may be, I think we have merch actually.
Do not affirm, do not comply.
Right?
Like you, the, the mama bear instinct has got to kick in here.
Yeah.
Uh, you know, do not, do not affirm your, child's decision that he's a tractor any more than his decision that he's a princess, and did not comply with reprehensible pharma executives' overrunning of the federal government to get you to put experimental shots into your children.
Right?
Do not affirm this thing just because they've used the word affirm and that sounds nice and positive.
I mean it's like the very language has been used to trick the more typically agreeable nature of women into doing this thing that is bad for their children.
Yeah.
The idea of believe your children when they say stuff, when in fact the point is, look, actually parenting, that ain't exactly new either, right?
The whole job of a parent is to affirm that part of what your kid is doing that should be affirmed and to steer them in the right direction for all of the stuff they shouldn't be doing.
That's what childhood is about.
And the whole idea that you're supposed to subordinate your judgment as a parent to what your child says doesn't make any sense at all.
Well, the child knows best.
Since when?
When did that show up?
There's plenty of bad parenting out there, but it's an insane position.
Yeah, it really is.
Maybe I'm done ranting for the moment.
We, I think, are done for this part.
I was going to say the first hour, but it's more than two at this point, I think.
And we're going to take a 15-minute break, and we'll be back.
We will be back with a Q&A today.
You can ask questions at darkhorsesubmissions.com.
We will start, as we always do, with a question from the Discord server, where they vote the people on the Discord server, which you can access at either our Patreons.
Vote on the question they most want us to answer every week that we do a Q&A, and we start there.
We also have at my Patreon tomorrow our private Q&A, so if you want more, join us there at 11 a.m.
Pacific.
And anything else to say before we sign off?
Nope.
I think we're ready.
All right.
Until we see you next time, be good to the ones you love, eat good food, and get outside.
