Symmetry, beauty, and the shape of consciousness

Transcript

00:00:00A humanity that's not growing is dying. The possibility space is really huge for both good scenarios and bad scenarios. We're human. This has good things and bad things about it. Should we count our blessings or should we be very eager to sort of strike out on adventure and explore?

00:00:19I do think that we're better than we realize, stronger than we realize, more capable, more beautiful, and more close to the good.

00:00:29I wouldn't want to just roll the dice and get a random mind from mind space, we can say.

00:00:35I think that the sky's the limit in terms of what the future of sensation could be. Human

00:00:41biochemistry being a little corner of this, and each person's qualia being a little corner of

00:00:49that. If things go wildly good, how good is that? If things go wildly bad, how bad is that? If we all

00:00:56die out, how bad is that? I think that the answers in all these cases are many orders of magnitude.

00:01:10This is Daniel Fagella. You're tuned into The Trajectory, and this is our 17th episode

00:01:16of the Worthy Successors series. Our guest this week

00:01:19is none other than neuroscientist and philosopher Michael Johnson. Very few people have thought

00:01:25hard enough and taken enough of a multidisciplinary approach to the problem of consciousness to maybe

00:01:32be able to scratch the surface of starting points to actually test for such a thing. Michael, I think,

00:01:37is one of the few who has. Obviously a very hard problem. He doesn't suspect that he's cracked the

00:01:43nut necessarily, but has some great ideas about where the future of sentience research might take

00:01:49us and the predominance and kind of preeminent importance of sentience in the forms of future

00:01:54life that carry on beyond man. I have a lot of thoughts for the end of this episode. As usual,

00:01:59I will save them for the end of this episode. Without further ado, this is Michael Johnson

00:02:04here in The Trajectory. So Mike, welcome to the show. Thanks. Glad to be here. Absolutely, brother.

00:02:11And we've got

00:02:13so many questions that we've got to answer. So I'm going to start with the first one.

00:02:13So much to fly into with the worthy successor theme. You were there from event number one out

00:02:18in San Francisco, a key thinker in this space, in my opinion. I know how I describe you to the

00:02:24audience, but I wondered if you could just kind of nutshell the nature of your research and core

00:02:29interests and sort of how you got into that just for people to kind of tee up where you're coming

00:02:34from from this perspective here. Yeah, yeah, that's a good question. So I describe myself as a

00:02:43neuroscientist. I do sort of work on sort of both how to turn consciousness into a real science,

00:02:51we can say, you know, if we if we look at the, the alchemy to chemistry transition,

00:02:56how do we do that for science of mind? And I think I would describe that I just,

00:03:04I'm so interested in this topic, because I think minds are so interesting. And so so centrally

00:03:11important.

00:03:13Of course, everything that we experience is sort of through the lens of our minds.

00:03:18Mediated. Yes.

00:03:20Yeah, yeah. And I also think that, like, I mean, I, my original intention was to move to the Bay

00:03:30Area and do AI safety. And I very quickly understood that I think about minds very differently than

00:03:41like the less wrong Mary.

00:03:43And, you know, I'm glad they're, they're doing their research. And I felt that, sort of, you know, I

00:03:55sort of tried to extrapolate, you know, with this kind of conception of mind, where do you get to?

00:04:02Do you get to a place where we sort of have a usable, actionable theory of moral value and

00:04:09consciousness and sort of navigation grade?

00:04:13Is that the best theory of mind? We can say. And my conclusion was no. So I kind of branched off and

00:04:21worked on more sort of formalism centric theory of mind. So since then, you know, been focusing on

00:04:31both sort of metaphysics of consciousness. What is the mind and also human neuroscience that how

00:04:40like, what kind of system are we?

00:04:41Yeah.

00:04:43You brought up kind of a distinction of moving out to work on AI safety and then sort of realizing some core differences that might not arrive at maybe a tight enough conception of what a mind is, a usable theory of sort of what the good is and how to navigate and expand it or even

00:05:00understand it.

00:05:01What do you describe those core differences as?

00:05:03Many people tuned in have seen the Yudkowsky episode or are certainly familiar with a Miri-esque conception of what AGI is and what kind of good and bad futures are.

00:05:12What did you see as sort of the core differences in the way you saw the world that you think kind of carved out a path that you decided would have to sort of be your own?

00:05:23Yeah. So, there was a wonderful little essay on Lesserong kind of explaining the Eliezer view, I would say.

00:05:34And it's that qualia is what an algorithm feels like from the inside.

00:05:40And it's sort of this very computational-centric perspective about sort of what a mind is.

00:05:47And a mind is sort of, you know, there's this computational system.

00:05:52There's maybe reflexivity, self-reference, and at a certain level of awareness, you sort of bootstrap into this metacognitive capacity that we call consciousness.

00:06:03That's not my perspective.

00:06:05I think that, you know, I sort of have claimed that there's really a question of whether we sort of associate minds or consciousness with software or hardware.

00:06:21Yeah.

00:06:23or bits. If we want to understand the consciousness of the system, the quality of the system,

00:06:28do we look at what the system's bits are doing, which computations are happening,

00:06:33or do we look at what the atoms of the system are doing, sort of what the physical configuration

00:06:40is? And my particular, I guess, flavor of this is I think that physical stuff is more real than

00:06:53computation, or more real than Turing-level computation, and that the sort of correct lens

00:07:00to apply to consciousness research is this lens of formalism. This idea that the proper goal of

00:07:11research into consciousness is to be able to create mathematical representations of moments

00:07:19of experience. And then, of course, you know,

00:07:23this kind of leads down a very different track than the former.

00:07:28And how do you sort of describe, like, some people would say, just hearing this the first time,

00:07:34some people would say, well, all right, so, you know, Yudkowsky et al. seem to think that sort of

00:07:40it's computation that's doing the quote-unquote stuff that conjures forth the inner field of

00:07:45awareness, the movie in the head. And then Mike thinks it's more like physics. And, you know,

00:07:53a lot of people think it's more like physics. And I think that's a lot of the stuff that's

00:07:54even a guy like me, it's like, well, do we really know what computation looks like at the smallest

00:07:57level? Do we really even know if quarks are a real thing or just something we kind of, like,

00:08:02it's sort of like, it's mysterious at the bottom either way. Why would that set you on a different

00:08:06trajectory for how to think about the good, how to think about minds, et cetera? Like,

00:08:10clearly it has, right? And you've given this much more thought than most people. But what do you see

00:08:14as the crux at that juncture that has distinctively pointed you in the direction that you tend to head

00:08:21as opposed to that larger crowd? Yeah.

00:08:24So in computation, generally speaking, like when we say computation, we often mean

00:08:32Turing-level computations. So a computation can run on a Turing machine. And if it can run on

00:08:39one Turing machine, it can run on any Turing complete system. So then, like, if consciousness

00:08:47is sort of the right algorithm, the right computation, then, like, creating pain points,

00:08:54pain and pleasure is more, no more and no less than sort of writing the correct software that can

00:08:59then run anywhere in a substrate neutral manner. My sort of observation here is that the, we can

00:09:10think of, like, Turing-level computations as a certain domain. And then we can think of physics

00:09:17as a different domain. And at least as we speak about them today,

00:09:24and as we sort of use the concept today, these two domains are differently shaped.

00:09:30You can't sort of line them up very cleanly. There's no objective way to map between

00:09:38what the physical microstates of a system are and which computations it's performing,

00:09:44or, you know, vice versa. So it's like, if these two domains are sort of differently shaped,

00:09:50then we really have a choice between, okay,

00:09:55is consciousness and, you know, if consciousness is the home of value,

00:10:00then it becomes, is computation the home of value?

00:10:06And creating a bunch of value is sort of running the correct computations.

00:10:13Or if physics is the home of value, then is creating a lot of value just like arranging

00:10:19atoms and energy in the right configuration?

00:10:24So, I often don't try to argue to convince people.

00:10:30I think that mostly people aren't argued into a theory of consciousness, so it's hard to

00:10:38argue them out of it.

00:10:39But I do think that like just kind of laying out, yeah, this is what I think is valuable.

00:10:48Yeah.

00:10:49Well, we're going to start to steer into where that takes you.

00:10:53I think that you're going to start to take a different direction in terms of what you'd

00:10:56want the future to be.

00:10:58Yudkowsky certainly has a vision of sort of how eventually the universe could wake up

00:11:01and be full of good stuff.

00:11:03A rather cautious vision, sort of famously so.

00:11:07Yours is quite different.

00:11:09So, we'll make our way into our kind of the main meat and potatoes where I think there's

00:11:14going to be some interesting points of agreement or points of questioning, all sorts of curiosities

00:11:21emerging.

00:11:21And I'll ask it this way and just let you roll it out the way you'd like to, is like,

00:11:27let's say however many years, a thousand, a million into the future, through waves of

00:11:33creative destruction that have, you know, kind of transformed everything that's ever

00:11:37come into being, we either were transformed or attenuated.

00:11:40And so, there aren't really opposable thumbs in English language like we have now.

00:11:43But if you could sort of somehow still observe this universe operating and see life continuing

00:11:51to flourish by the definition you would distinguish flourishing, what would be going on?

00:11:56What would be the traits of that intelligence or intelligences?

00:12:00What would be occurring in that space for you to say, you know what?

00:12:03Okay, I guess there's no more of my direct descendants, but this really feels like a

00:12:07win.

00:12:08What would that be for you?

00:12:09So that's kind of the central question for philosophy.

00:12:13And I guess I would reframe this as what is the good?

00:12:19And how would we know?

00:12:20Okay.

00:12:21How can we live in a universe with a lot of it?

00:12:23How can we choose paths with more of it?

00:12:26And is there sort of one unitary, you know, philosophers talk about the good in, you know,

00:12:32capital T, capital G.

00:12:34Yes.

00:12:36And then there's also this perspective that a lot of it is about preference.

00:12:40Like I like this, I don't like that.

00:12:42Maybe you have different preferences.

00:12:44How do we navigate that?

00:12:46One of my big sort of...

00:12:50Okay.

00:12:51Departures from Yudkowsky is that I think that the good has a unitary nature.

00:13:01And this actually comes from looking at neuroscience.

00:13:06For example, I think Barrage and Kringlebach have an observation that, you know, kind of

00:13:13reviewing different hedonic studies about pleasure in the brain, and they say that essentially,

00:13:21you know, whether you're eating a candy bar or listening to a transcendent symphony, the

00:13:33same regions of your brain sort of light up on fMRI.

00:13:38This isn't proof that, you know, the same thing is happening in your brain, but it is

00:13:43suggestive evidence.

00:13:45And I think that essentially, like, I think that we all can have different preferences,

00:13:51but I think there's some sort of unitary success condition.

00:13:57And so this gets into my symmetry theory, valence theory.

00:14:03And essentially, you know, to sort of look at your question directly, we were just talking

00:14:13before the show about Giulia Tononi's idea, and I think you've also given a TED talk on

00:14:19this idea.

00:14:21Yeah, yeah.

00:14:21About, yeah, about this idea of a qualia scope.

00:14:25And basically, you know, the sort of way, you know, lens that you can kind of look through

00:14:31and look around, and it can see consciousness, it can sort of see experience.

00:14:38And so, you know, maybe you look over at, you know, your neighbor and you see a big

00:14:44glow.

00:14:44And, you know, maybe you can tell that he's having a good time.

00:14:49Maybe you look at a...

00:14:51Your pet dog and he, you know, there's a big glow there.

00:14:54He's having...

00:14:55He's also...

00:14:55He's a golden retriever, so he's super happy.

00:14:58Yeah, yeah, yeah.

00:14:59Always having a good time.

00:15:00Yeah.

00:15:00Yeah, yeah, yeah. And so, I guess, you know, to understand whether a future is good

00:15:09in a valence sense, and you know, I'm, I tend to be like ontologically utilitarian,

00:15:18but practically virtue ethicist. So, you know, I don't want to over index on utilitarian.

00:15:23Yeah, yeah, yeah.

00:15:24But, yeah, yeah. But I mean, I would want to look around in the future and see, okay,

00:15:32how much consciousness is there? And are people having a good time? And I mean, this kind

00:15:38of connects to, for example, Nick Bostrom has this famous quip that Disneyland with

00:15:44no children has no value. And I think there's something really important there. And so,

00:15:50in a good future, I would want, first of all, there to be consciousness, qualia. It's like,

00:16:00it's the substrate of value. It has to be present for there to be positive or negative

00:16:06value. And then I'd want a lot of positive value. And I think the, like, we can say the

00:16:12game theory of like, what systems work to produce a lot of positive value, you know,

00:16:17that's another question.

00:16:19Yeah.

00:16:20And then also, I think that just sort of, I wouldn't want stasis. I think that there

00:16:26are many local minima we could get trapped in along the way. And I want to make it all

00:16:32the way. Eventually, we don't have to be in a rush. But something that permanently kind

00:16:39of takes better futures off the table, I think would be negative.

00:16:46Yeah, let's, let's double click on both of those. My goal here is really clarifying,

00:16:50clarifying the worthy successor vision in the sense of, if we aren't around, what would

00:16:57still be a splendid victory for the great process of life of which we are part? So my

00:17:01goal is to clarify that in great detail with you. And then like, poke at the edges, we're

00:17:07like, I still have a lot of questions about how you think about it. And like, just look

00:17:11at different perspectives here. So one of them is, you know, a very rich, sentient future,

00:17:17with preferably kind of passive, passive, passive, passive, passive, passive, passive,

00:17:20positive quality, like states, or quality of states of different kinds. And you had

00:17:25said kind of, I guess people having a good time in this particular case, I guess you

00:17:29mean the total space of minds, agents, etc, that could have such an experience, right?

00:17:36Because it may not be up, you know, people in the current sense, but do you suspect that

00:17:44that could include, like, in terms of the panoply of what that could include? I think

00:17:48for some people, it's like, well,

00:17:49each robot has its own little mind and like those things, you know, and then for other people, I think it's

00:17:54I don't know, maybe we've got tiny little sugar cubes that do some like weird collective

00:17:59intelligence thing, but they all have their own little thing. And then maybe we have,

00:18:03you know, Jupiter brains. And then maybe in some sense, maybe there is, you know, when

00:18:07you talked about the qualia scope, maybe you look at an ecosystem and the whole ecosystem

00:18:11actually like you and I are, you know, we can't know if our liver is minorly sentient

00:18:18in some regard, because we don't, it seems like, you know, we're not, you know, we're,

00:18:20like, from our vantage point, we don't get access there. But it seems conceivable that

00:18:25we don't really know where sentience actually sits. Like, when you say it or persons, give

00:18:31us the range that you suspect that would include in a huge ecosystem of post human minds.

00:18:38Yeah, yeah. So it's a good question. And I would also say that, you know, we can also

00:18:43give us or give the range that we might expect to see on the earth now.

00:18:49Yeah, sure, sure.

00:18:50Yeah.

00:18:50Go ahead.

00:18:51Yeah, yeah. Hundreds of years ago, you know, apparently Leonardo was dissecting dogs without,

00:18:59you know, thinking that they could feel and I think that, you know, that was not so good.

00:19:04And like, I think it's cool that now we like, see animals as Yeah, like, they're also creatures

00:19:10that can, can feel good and feel bad. And like that matters. I think that, you know,

00:19:18I have some, some work that kind of digs into that.

00:19:21what's out there and some others.

00:19:25And it's this idea that any future theory of consciousness,

00:19:31if it's really like,

00:19:32and like the vision is having a theory where you can literally just shut up

00:19:39and calculate, it'll figure everything out. You know, it's,

00:19:43there's no subjective finger on the scale.

00:19:46It's just math and you can derive what is conscious,

00:19:51how it feels, et cetera. And I actually think that it's,

00:19:54it's more tractable than one would realize.

00:19:56So we can talk about dual aspect.

00:19:58We're going to get a lot into that. It turns out

00:20:00how to measure it, we got a whole section, so we will be deep diving, sir.

00:20:03Nice, nice. Yeah, so like one interesting proxy, and you know, we can talk about like,

00:20:11just like generation of qualia in general, or generation of like big sort of chunks of bound

00:20:19qualia, we can say. Like, you know, we, our minds have like a bunch of different things kind of

00:20:25glued together into this, you know, crazy, crazy experience that we have. You know, here's some,

00:20:33some blueness, some redness, some heat, etc. So like, the conditions for consciousness versus

00:20:41the conditions for binding can be interestingly different. But, you know, looking at the world

00:20:49today, I think, yeah, of course, humans are conscious. And yeah, pretty obviously, animals

00:20:55are also conscious. Frankly, they could be even more conscious than humans are. So that's like,

00:21:04an interesting digression. Kind of, you talk about like, blue, blue whale, blue whale, brain size,

00:21:10dolphin brain size, octopi, you know, little brains all the way down the tentacle, like, when

00:21:18you say more conscious, just, just to put a sentence on that caveat, what do you mean there?

00:21:23Yeah, so,

00:21:25like, Eleven very, very clearly spells out, like, a lot of, a lot more things could be minds than we

00:21:34think in TAME. Oh, yeah.

00:21:36Teleological approach to minds everywhere. Really, really great frame. I think that also that,

00:21:44like, perhaps animals could have, could sort of shine more brightly than humans,

00:21:50in some cases, at least, that, and this gets into the the vessel computation topic.

00:21:55We were chatting about it a little bit. And this idea that maybe animals are, by default,

00:22:02more in a state of persistent, non-symbolic experience, sort of a non-dual state, you can say.

00:22:10And so, you know, a deer in the forest, you know, maybe like, that deer feels like the

00:22:17sweetness of life and the pain of life more intensely than humans who have a lot of frozen

00:22:23sensations in our, you know, the environment. Yeah.

00:22:25In our minds, we can say. So, yeah, that's a big rabbit hole.

00:22:31We're gonna get there, brother. We're gonna get there.

00:22:33Yeah.

00:22:33It's good to know what you mean by more conscious in some respect. Hopefully,

00:22:37now the audience is following you. So, that makes sense.

00:22:39Yeah, yeah, yeah. Yeah. And I mean, it's like, philosophy is gonna be weird. It's not gonna,

00:22:45like, check all our intuitive boxes about, you know, how we kind of think things work. And so,

00:22:53um,

00:22:54if we did have this qualia scope where we could see conscious experiences, you know, we might look

00:23:00at another person. Okay, there's consciousness there. A dog, there's consciousness there. A deer,

00:23:04oh, wow, that's really bright consciousness. Insects, sure, consciousness. But what about

00:23:13data centers? Sure. I think, like, there's something really interesting going on there

00:23:19in terms of, like, physics.

00:23:23Yeah.

00:23:24And it's somewhat heterogeneous, actually. And, you know, even,

00:23:30like, a power transformer or a particle accelerator, nuclear reactor, etc.

00:23:37There's, like, something really interesting and extreme in physics going on in all of

00:23:44these examples where, at least from some proper theories of consciousness that are fully objective,

00:23:52you can shut up and calculate. Yeah.

00:23:54there could be extreme amounts of consciousness and or valence there so you know i i like to say

00:24:02that um you know i wish i lived in a world where people were protesting nuclear nuclear reactors

00:24:09um and like not because of some environmental thing you know okay maybe there's there's like

00:24:16arguments on both sides i'm probably pro-nuclear uh but because some people thought that well

00:24:24you're creating a lot of consciousness and it feels really terrible and so it's like worth

00:24:29protesting and then there could be counter protesters which say you know yeah it's you know

00:24:34the these human-made structures are creating a lot of qualia but it's really good so we should

00:24:41make more of them and uh there's just so much richness in

00:24:47sort of the space of what is probably happening already that is invisible to us uh and i think

00:24:54that yes often people just kind of over index on things that we know

00:25:00to the exclusion of things like that are pretty plausible if you just kind of look at the math.

00:25:11Yeah, well, I mean, Levin talks about, you know, we're used to medium-sized things moving at medium speeds, and those are all the things that matter, and those are the things that could be conscious, and da-da-da-da-da.

00:25:20And, you know, to your point, if we think about what sentient things have in common, the scope could be larger.

00:25:28I'm going to zoom this into—I'm going to ask a series of questions just to flesh out where you see the bounding box on this and just kind of click in,

00:25:36because have people get a better conception of what it means for good qualia to bloom in the way that you would see as a win for the great process of life of which we're part.

00:25:45So, you were like, hey, is it possible?

00:25:50Does a nuclear reactor feel something?

00:25:52You know, does a sun feel something?

00:25:54Does a hurricane feel something?

00:25:57You know, we could wonder, again, does each individual kidney feel something?

00:26:02We could ask, does the city of Austin feel something?

00:26:06And if that's the case, well, surely then there's some Gaic meta-sentience around maybe the entire planet in some respect.

00:26:15And if that be the case, then maybe already that's, you know, there's a thing.

00:26:19There's a feeling like aScript.

00:26:20the Milky Way, which might be outlandishly more granularly interesting than what we could feel.

00:26:25I can also see that not being the case, by the way. I'm not positing these things are true. I'm

00:26:29just saying we don't have access necessarily. Are you seeing all those scales up, scales down,

00:26:36different shapes, different mediums, all potentially being sentient if we could

00:26:40sort of understand them? Or is there anything you want to add to that?

00:26:44Yeah. I don't think they're ontologically equivalent, and this is important.

00:26:51But basically, I think some levels of abstraction are more real than others.

00:26:57And consciousness lives at the most real level of abstraction. And so if we talk about, you know,

00:27:03is the city of Austin conscious? Is Walmart Corporation conscious? Is like the World

00:27:09Series of Baseball conscious? My answer would be no, definitely not.

00:27:14Yeah, yeah.

00:27:15And so I see consciousness as being deeply linked to a particular

00:27:20chunk of space-time, probably a contiguous chunk in some way. And so, you know, if we looked at the,

00:27:29you know, through the qualia scope, every sensation would be sort of anchored in a place.

00:27:35And like, you know, there would be maybe like these bubbles of consciousness where

00:27:43the different sensations within that boundary are all linked together.

00:27:48But I think that,

00:27:50some ways of speaking about systems are leaky reifications, we can say. And I think that

00:28:01these leaky reifications aren't conscious. That I don't think postmodern society is conscious.

00:28:11I think that if we point to like a specific place in space-time, then we can talk about consciousness.

00:28:18But if we can't, we should find an alternate level of reality to speak about.

00:28:24Which gets into this question of AI consciousness. And I wouldn't say.

00:28:30Yeah, so yea, it's going.

00:28:31Yeah, yea yea. I mean, one of the most interesting questions today. And I do strongly believe in

00:28:37being polite and honest with models. I don't think that CLODD is conscious. I don't think that GPT-5is conscious.

00:28:48and so on. But I think the servers they're running on are physical and there could be

00:28:54quality associated with those. So I'm going to like paint out like one of the big challenges

00:29:02for philosophy over the next decades is trying to construct a working self-report model for

00:29:14AI consciousness. Right now we can ask Claude, you know, do you feel hungry? And like, you know,

00:29:20with enough prompting, we can say, we can get Claude to say, yeah, like I feel hungry and my

00:29:26favorite meal is like pizza and whatnot. And like, do we believe it? No, I don't believe it.

00:29:32But like, I want that for AIs. I want sort of whatever

00:29:41sort of fragments of qualia that are sort of like, you know, like, you know, like, you know,

00:29:44associated with, you know, their physical extension in data centers or whatever,

00:29:49to eventually be connected with their qualia reports.

00:29:54And so, you know, that's a, that's a big research paradigm.

00:30:00just kind of like how do you finally sample the physical world in a way that would let you know, for example, what's going on in the EM field or brain, local brain field space or whatnot, and then connect that to like a self-report paradigm.

00:30:17Certainly.

00:30:18I mean, it seems as though to have the kind of flourishing of qualia that you're talking about, it would need to become substrate independent at some point.

00:30:28Right?

00:30:29Guanine as the basis of all sentience would seem actually pretty limiting for sort of qualia writ large.

00:30:36And it does seem unlikely that qualia should be married to Guanine for eternity.

00:30:40That there be no other way of organizing atoms, I guess, in your paradigm to conjure forth the field of awareness that you and I presumably are experiencing.

00:30:52So what is the strongest thing?

00:30:58I think it was Peter Voss leading up to this interview for the interview with you.

00:31:02What is it that mostly leads you to the conclusion that AIs likely would be able to be conscious and even volitional?

00:31:10You know, you also talked about no stasis, kind of this growth, et cetera.

00:31:14We're going to get into that too.

00:31:15We're mostly focusing on qualia here.

00:31:16But what primarily leads you to that supposition that eventually we will see a kind of substrate independence for this kind of evolution?

00:31:28What is the kind of mind that you're studying?

00:31:30Yeah.

00:31:31Well, I want to be careful about the term substrate independence.

00:31:35I think there are, I would say, there are naive ways to implement it that don't transfer qualia.

00:31:42And then there are more sophisticated ways that would.

00:31:45So, for example, you know, there are ways, you know, we can talk about mind uploading.

00:31:49And I think there's a naive way to do it that would sort of make you into a non-sentient chatbot.

00:31:55There wouldn't be any experience that would sort of track how you described how you felt.

00:32:06And then there would be a more sophisticated sort of hardware and software approach that would actually replicate qualia.

00:32:12And, you know, a very simple story there would be if we replicate the EM field of the brain, we may replicate its qualia.

00:32:23So not the rabbit hole.

00:32:25There it goes pretty deep.

00:32:30Yeah.

00:32:31Just your intuition in general that it is possible for, I guess, you know, I certainly don't, I'm not making the argument by any means.

00:32:38You're bringing up a good distinction that not every way you could do things that seem to do what we associate with minds would automatically have an inner field of awareness.

00:32:48I couldn't agree more there.

00:32:50But I guess your position is it doesn't have to be probably.

00:32:54It doesn't have to be by chance.

00:32:56you would suspect it seems that at some point something that we would now think of as a

00:33:03computer or machine would be able to experience qualia.

00:33:07Some people do not share this intuition at all, or maybe some people haven't given it

00:33:11as much thought or have come to different conclusions.

00:33:13But what has led you primarily to believe that machines will be part of the flourishing

00:33:17and expanding of mind space?

00:33:19Where does that, what has led you to that conclusion primarily?

00:33:23Yeah.

00:33:24So first, I think it's important to sort of distinguish different definitions of consciousness.

00:33:31And I'm talking about sort of phenomenological consciousness.

00:33:35So like what it feels like to be me right now, what it feels like to be you and so on.

00:33:40So not like access consciousness or awareness or whatnot.

00:33:44But I think the basic argument is that we're made of the same stuff.

00:33:51It's like I'm made out of like.

00:33:54A lot of the same kinds of atoms as my phone to say that consciousness is like forever biological

00:34:04and that we couldn't like technologically recreate it.

00:34:10I just don't think it makes ontological sense.

00:34:14I don't think you can kind of have a full theory of consciousness with a special carve out for and

00:34:21only humans can be conscious.

00:34:23Right.

00:34:24Or only biological creatures can be conscious.

00:34:26Bio life.

00:34:26Yeah.

00:34:27But I guess like I'm, I, I love Michael.

00:34:31So I want to like offer full, full respect for his tame teleological approach to mind everywhere approach.

00:34:43I think of like I'm zeroing in on like phenomenological consciousness, like a certain definition of the mind where it's less sort of.

00:34:54You know, intelligent system and more chunks of sensation.

00:35:00and I think these chunks of sensation are very bound to what's happening in the physical world.

00:35:08Yeah, and so, with that paradigm, we would suspect then that chunks of sensation could be manifested in physical ways that might not involve guanine,

00:35:18but would nonetheless have something that it phenomenologically feels like to be that thing or that, you know, entity, agent, chunk in some way, shape, or form.

00:35:29Okay, got it.

00:35:30So, in terms of the grand expanse of this, some people presume, well, the grand expanse of consciousness will be more of the things we understand,

00:35:39more humor, more love, more chocolate ice cream, maybe a flavor of ice cream we've never even tasted before.

00:35:45Ooh, that would be really cool.

00:35:46And it's kind of like an exaggerated version of what hominids know now, kind of like the Greeks would say,

00:35:50yes, well, you know, the gods, they ride a chariot in the sky, and they have a super sword that, you know, is like indestructible.

00:35:57It's just like extended analogies of like monkey stuff.

00:36:00When you say this kind of grand expanse of sort of quality you would hope to blossom and bloom as kind of a good future, I think you mean something different.

00:36:09Can you articulate to us, is this still bound to the pain-pleasure axis?

00:36:13Is this still bound to the comfort?

00:36:16Are there any kinds of emotional states that hominids experience just in greater extents?

00:36:20What do you see as that space, and how wide should our imaginations be?

00:36:24Because I think most people don't have one quite as wide as yours.

00:36:28Yeah.

00:36:29Yeah, it's a good question.

00:36:32I think that, you know, there are as many—definitionally speaking, there are as many flavors of qualia as there are different states and arrangements and dynamics of matter.

00:36:47So that's, you know, much, much bigger than just the human space.

00:36:53And I do think that, like, the question is always, okay, well, how do you put constraints on the topic?

00:37:01That constraints allow us clarity about what can be, what can't be, and sort of allow us to sort of build up an understanding of the actual possibility space.

00:37:13And so I think of this in terms of what are the natural kinds of consciousness?

00:37:25What are the natural kinds of qualia?

00:37:27And, like, if you're sort of building—trying to build a new science, if you're trying to sort of systematize this idea of qualia, then you really do need to be opinionated about, okay, what are the natural kinds?

00:37:42And, you know, that's—

00:37:45One—I have a couple threads here.

00:37:47One being, I think emotional valence is a natural kind, that it's well-defined across all conscious experiences.

00:37:58Where it's like we may discover new, like, dimensions of variance in this, but at the same time, it's going to be this unitary thing that, you know, a million years in the future,

00:38:13we're going to be able to see that the human successors that we have will still be somewhere on the valence spectrum.

00:38:18And this will still be important.

00:38:21Just from a, like, you know, not throwing shade on their technological prowess, but just as an outgrowth of, like, yeah, this is like the laws of physics.

00:38:33It's the laws of qualia.

00:38:35And, like, you can't break the law.

00:38:38You can be clever about, like, how to navigate that.

00:38:45And then, I guess, secondly, just, like, thinking about kind of the characteristic taste of human qualia and, you know, how big of a possibility space there is that's outside of this, you know, what we can feel.

00:39:05I do think, like, one of the things that I'm kind of looking into right now is what I would say is cellular qualia.

00:39:12Sort of—

00:39:13Yeah.

00:39:14—are like cells as minds.

00:39:16And they had these evolutionary imperatives to sort of navigate the world and so on.

00:39:22And perhaps, like, our sensations are built on the sorts of situations our cells had to navigate millions of years ago.

00:39:34And so if there's, like, a spill of some sour chemical, like, get away and, like, cringe from that.

00:39:41And, like, our sour tasting cells will be like, oh, this is what I'm going to be like.

00:39:43buds are sort of built off that response. Likewise with, so, you know, if the cell

00:39:50senses a lot of hydrogen ions, it should like symmetry break in a certain way.

00:39:57Whereas if the cell senses

00:40:00a lot of amino acids with hydrophobic side chains, so bitter sensations, it should sort of symmetry break in a different way.

00:40:08So, you know, there's a quip that human biochemistry is water chemistry.

00:40:20And I think that's more true than people realize.

00:40:23We sort of, there's a famous quip about how we've sort of taken the ocean with us in ourselves.

00:40:33Sort of, you know, based on software.

00:40:35Yeah, yeah, yeah.

00:40:36Yeah, yeah.

00:40:36And I think that it may turn out to be that human qualia is water qualia more so than anyone has realized.

00:40:50We're going back to the world composed of things.

00:40:53There's elements in the ancient Greeks and Chinese here.

00:40:55You're like, we're like water spirits and there might be like fire spirits.

00:40:59Yeah.

00:41:00No, but I see what you mean here.

00:41:02So you're saying the space that we're sampling is bound from sort of the trajectory we began from.

00:41:12And so there could be, you call it natural kinds.

00:41:16You know, to you, Valence, pain-pleasure axis is I think what David Pierce often uses.

00:41:22Yeah.

00:41:23Is to you sort of a sort of natural kind.

00:41:25In other words, when sentience be there, that be one of the things that be in sentience.

00:41:30But you're also saying we are sampling a space of the bubbling up of possible minds from the archaea from which you and I began as.

00:41:40And then the flatworms and then the pre-reptiles and then eventually reptiles and mammalian, whatever.

00:41:47So we're, yeah, we're kind of carrying in some sense like a lot of extras.

00:41:53On top of like the way the very earliest ancestors had to respond to the world.

00:41:59And that this great bubbled up place, this vantage point we have is still, you know, we could say it's empowered by.

00:42:08We could also say it's fettered by our origins.

00:42:12And so do you suspect the set of natural kinds within qualia?

00:42:18It sounds like when you said it's as varied as the ways we could organize matter or what have you.

00:42:22Yeah.

00:42:23That sounds to me like there could be a near infinite number of natural kinds.

00:42:27In other words, pain pleasure is something you and I can experience.

00:42:30But there may be like another axis or another strata or another set of feelings or another way of experiencing time.

00:42:38You and I have this sort of like frame after frame.

00:42:41I could imagine something that's playing at any given time, you know, eight different inner films.

00:42:48Some going forward fast or some going forward slow.

00:42:50Some going backwards.

00:42:52Some almost.

00:42:53And it's like a process of ossifying the present.

00:42:54And it almost feels like the matrix to that entity and whatnot.

00:42:57And so do you have a suspicion of what some of the natural kinds could be hypothetically?

00:43:07You and I maybe can't access them.

00:43:09But people would need to be like, well, what could he mean?

00:43:12Like if something bubbled up from non-water on some faraway galaxy somewhere.

00:43:18And it was sampling a different space of this like what it's like to be somewhere.

00:43:25What do you suspect could be examples of other natural kinds of the kind, Michael, that might flourish in this world of minds beyond our own?

00:43:34Like the kind that maybe you would hope we get to explore and to see that would unravel new powers and experience.

00:43:39Like give us your thoughts on that.

00:43:41Yeah.

00:43:42That's a great question.

00:43:45I mean, I do think that, you know, to understand human qualia, you know, we should be looking at cells more than anything else.

00:43:53And sort of different.

00:43:55Sort of programs that cells had to run to deal with this or deal with that or whatnot.

00:44:01And sort of cells are very smart.

00:44:05I think, you know, once again, Mike 11 is under something there.

00:44:09So, you know, just kind of if we want to know like the human state space, look at cells.

00:44:17And that's what I'm looking at right now.

00:44:21But I think that, you know, maybe in.

00:44:25Halfway across the galaxy, there was another planet and it was just made of like liquid mercury or something, you know, just like and boron and argon and.

00:44:35Yeah, whatever.

00:44:36Yeah.

00:44:37Lithium.

00:44:37Yeah.

00:44:38And some sort of like alien xenobiology evolved using those elements.

00:44:45And well, that would have like a completely different basis set of basic sensations, you know, instead of like sour and heat.

00:44:55And whatnot.

00:44:56You know, it would have blorg and zorp and, you

00:45:00sorts of stuff. And so, yeah, and one really important thing there is it would also have a default, like a different default, like phenomenological default hum, which would be a different default valence.

00:45:24Like, it could just be like radically better to have that biochemistry or radically worse. And this again will eventually become a quantitative question.

00:45:37But closer to home, you know, we can look at, we can look at computers and we can, you know, ask, okay, what's going on in there?

00:45:44Like, this is, we really, we kind of have alien biochemistry already.

00:45:53It's a little bit different than self-organized biochemistry.

00:45:56So, like, biology is very much, it sort of emerged from feedback cycles and epicycles and replication and whatnot.

00:46:08So, it has a certain profile and like runs on like certain distributed algorithms, we can say.

00:46:17And whether we call that free energy minimizing, whether we call that distributed algorithms,

00:46:25or just kind of thermodynamic dynamics.

00:46:31Self-organized systems will have a certain kind of flavor and computers are unlikely to have that flavor.

00:46:40Instead, like they're sort of buffered against noise.

00:46:44It's called Johnson noise, the thermal noise of the circuit.

00:46:48No relation.

00:46:50And I think that like, you know, I'm, you know, I'm not a big fan of the term, you know,

00:46:55but in my technical consciousness research, I think a lot about sort of the branchial shape of different systems.

00:47:02And what is the branchial shape?

00:47:04Kind of, Stephen Wolfram has this branchial space construction where for every sort of branching interaction,

00:47:12where like the molecule, like the cesium atom decayed randomly.

00:47:18And so we can kind of model it as if, okay, in this universe it decayed,

00:47:23in an alternate universe it didn't.

00:47:25And just like make a graph of all possible random quantum events.

00:47:31And then I'm saying that like, if we did that for the brain, we'd get like a pretty cool shape,

00:47:37a lot cooler than like a chair, for example.

00:47:41But then if we do that for a computer, we get a very different kind of shape.

00:47:46And I guess I think, you know, that that's significant for digging into computer consciousness.

00:47:56I'm a lot more comfortable saying machine consciousness than I am AI consciousness,

00:48:03because AI is more software.

00:48:05I think that this idea that different shapes have different,

00:48:09or different objects have different shapes in branchial space is very important.

00:48:13And, you know, I like to think of a true shape lives in branchial space.

00:48:20And, you know, if we want to sort of understand an object's true nature,

00:48:23we should sort of expand it into something else.

00:48:25And then we can look at the shape of the object in branchial space

00:48:25and see sort of the branching decoherence of that object.

00:48:31And likewise, I think minds also live in branchial space.

00:48:36So, you know, if we figure out a boundary condition in branchial space,

00:48:39then we can kind of really see the shape of an experience

00:48:44in this sort of Stephen Wolfram's branchial space construction.

00:48:50And I guess, I mean, this really points me to like, you know,

00:48:54I were to think really deeply about machine consciousness,

00:48:58I think I'm more comfortable with machine consciousness than AI consciousness,

00:49:04because AI is sort of software and machine can be...

00:49:07Yes, yes, yes. I'm with you. Yeah, yeah, yeah.

00:49:09Yeah. Then, you know, I think a lot about branchial shape.

00:49:15I think a lot about binding.

00:49:17What is like the boundary in branchial space that might separate one experience from another.

00:49:23And then I think a lot about the idea of, you know,

00:49:26CMOS, like the actual physical substrate of the chip,

00:49:32the chemistry of CMOS that, you know,

00:49:38we have these incredible machines and like factories that do like super tiny details on chips.

00:49:47And we do very careful material science to sort of build out, you know,

00:49:55the

00:49:56the 3 nanometer process and the 2

00:50:00nanometer process coming and whatnot. And I think that the sort of chemical details of what the chip is made out of may actually lead to like, it's kind of the same substrate as like, okay, humans have cells, and like human qualia sort of emerges from like aggregating cellular qualia, we can say. And I think that

00:50:28machine qualia will arise from the substrate itself as well.

00:50:35Yes, it will, I guess your suspicion here is if we get closer to what sorts of constructions are capable of conjuring sentience, then we will have informed opinions about the materials with which we build and put things together. To your point, there may be some ways to kind of build things that can output mind like stuff, but which which

00:50:57don't house sentience. Just to make sure we touch on your second point, you talked about having a grand flourishing of this state space of minds, but that there are almost certainly bounds on what those minds are. They'd have to be built of certain things. They would open up new magazines of space, but those magazines of space might all have some commonalities. Maybe they all have pain, pleasure, et cetera. We're going to probably chip into some bits of that just to see where that lands. But you'd also talked about, you know, when you could look down and maybe there's no more humans, but there's minds experiencing rich and varied things, different kinds of experience, different kinds of powers, different kinds of spreading and activity and behavior in the world. Martin

00:51:40Martin Rees described it, and you may have a different opinion, but sort of like, as wildly complex as human civilization is to like an individual sea snail would be sort of what we could go down and look at, we would hope to for it to be a rich, varied ecology of which we probably don't really know what's going on, but but we at least know it's richly sentient, and it's doing its pursuing things that seem to be the best behooving of it. You can challenge that or add a conduit to that if you'd like. But one bit of that is your supposition that you would not want it to be static.

00:52:10now it seems to me on some level Mike that I don't know if there is static. In other words, I don't know if being is actually real. I actually it seems to me like becoming is real. And being is sort of like, at least from our temporal way of seeing the world. It almost I'm less certain about being. It seems like permanence is a matter of degrees. Stasis isn't even viable. When you say I wouldn't want stasis.

00:52:36What would you want? And what would that imply? Like the opening up of I think you would say,

00:52:40called it sort of like, you wouldn't want different futures to be taken off the table.

00:52:45You would want optionality to be kind of captured. Like why? Why is that morally important to you?

00:52:51I'm not challenging it, but I just want to make sure I understand what you mean. What does

00:52:55non-stasis, what does the exploring and not cutting off of optionality mean for you?

00:53:00Yeah. So I'll lead with a quick comment about, I think that humans are great. And I think that

00:53:10in ways that we don't even realize and that the universe will miss us when we're gone.

00:53:19And I think that like just understanding a little bit more about why humans are cool

00:53:27and what attributes that we have that future AIs might not get by default. I think that's

00:53:34going to be important. I also think that stasis, you know,

00:53:40in the grand sense of things, stasis is impossible. There is narrow time and so on.

00:53:47But at the same time, I could see practical situations arise where we would get a lot of

00:53:54like something we would call stasis and maybe permanent reduction in possibility space.

00:54:00For example, if this is essentially like a winner-take-all AI race and the winner gets there

00:54:08and then basically decides, oh, I'm going to win this race. I'm going to win this race. I'm going to win this race.

00:54:10okay, you know, I'm going to make sure that I always stay in power.

00:54:16And I'm just going to like, you know, have LLMs check on like everyone in the world.

00:54:23And anyone doing anything disruptive gets the ax or whatnot.

00:54:28I think that essentially we would get, we could get something like stasis.

00:54:34And I don't think that would be a good, good future.

00:54:37There are scenarios where the flame itself might still be bright because it's smart,

00:54:43but it's just focused on staying this small ember.

00:54:46It's almost like it's optimizing for embering.

00:54:49It's not a blaze into the multiverse.

00:54:51You're saying that there could be an initial set of goals that might like intentionally

00:54:57and maybe from an Earth life perspective,

00:55:00perpetually shrink or hamper the flame itself? Is this what you're getting at?

00:55:07I would say so. And I would also say that a humanity that's not growing is dying. So it wouldn't be stable in that sense either.

00:55:16Yeah.

00:55:18But I do think that the possibility space is really huge for both good scenarios

00:55:24and bad scenarios. I think that it's this question of Chesterton's fence and we're human.

00:55:36This has good things and bad things about it. And should we count our blessings or should

00:55:42we be very eager to sort of strike out on adventure and explore? I do think that we're

00:55:49better than we realize, both stronger than we realize, more capable, more beautiful,

00:55:59and more close to the good. And so I wouldn't want to just roll the dice and get a random

00:56:09mind from mind space, we can say.

00:56:13But I think that the sky's the limit in terms of what the future of sensation could be.

00:56:21And that human biochemistry being like a little corner of this and each person's qualia being

00:56:28like a little corner of even that. The possibility space is really vast. And really the core

00:56:36question in all of this is like, what's the potential upside?

00:56:41Yeah.

00:56:42Like, what's at stake? What's like, if things go wildly good, how good is that? If things

00:56:50go wildly bad, how bad is that? If we all die out, how bad is that? And I think that

00:56:59the answers in all these cases are many orders of magnitude. Like, I don't think it's just

00:57:05like, you know, okay, in the future, like everybody might be like 20 to 50% happier.

00:57:10No.

00:57:12A million times happier. So I think that like,

00:57:19hedonically speaking, a lot is at stake. But also just in terms of like, other virtues as well.

00:57:27I think that, yeah, it's hard to talk about future stuff through the lens of consciousness,

00:57:37because we don't have a canonical theory of consciousness yet.

00:57:42Yeah.

00:57:43But from everything I can see, the possible variance is just extreme.

00:57:52Yeah.

00:57:53Yeah.

00:57:54is why you would would not want stasis. You brought up an important point. So now I'm gonna

00:57:57Yeah.

00:57:57these two. points and we're going to get into measurement and a little bit of what you hope innovators and regulators do. You just said we shouldn't explode into some random part of mind space. That's probably not good. In fact, there's a lot of good in our own experience. I would

00:58:09concur with you i think we're we're about as bright a flint we're about as bright a torch as

00:58:14we have known of uh in terms of sentient richness in terms of our own self-creative ability we have

00:58:19this whole magazine of potential beyond biology of culture and then of technology and we've opened up

00:58:24entirely new categories of magazines of powers uh and acts and understanding of nature i mean

00:58:29baconian science etc like we're doing stuff that and and you know love humor etc by golly you know

00:58:36just rolling the dice again feels a little goofy so we're going to talk about the trajectory

00:58:41question that you just touched on but let me kind of almost question playfully jump into devil's

00:58:47advocate or in a couple things so one of them is um kind of uh qualia as sort of where the good

00:58:54cashes out on some level uh and you brought up a good point of like well if you know if the flame

00:59:02itself is snuffed out then there really isn't any room for the good if it's not

00:59:06growing it's dying sort of deal this this does seem to be the case and so it would seem to me

00:59:12as though if there be value there there is a value actually in the unraveling not not just

00:59:19more complexity more novelty i actually have like very little sympathy with that i think like what

00:59:26nature seems to have not done mike is like conjured forth weirdness and perpetuated pure

00:59:33novelty weirdness for the sake of

00:59:36pure novelty weirdness this actually strikes me as not the case what strikes me as the case is

00:59:41it has bubbled forth magazines of powers and access to nature and experience that have behooved the

00:59:47non-death of the thing and or the system that actually seems to be what's going on it's not

00:59:52oh it's it's bringing forth more novelty well if the novelty is just charity burdening charity

00:59:59weakness

01:00:00that like the part of the system that's actually productive has to hold up.

01:00:03At some point, the system itself crumbles and then you have no value.

01:00:06And how valuable was that?

01:00:07Because when you snuff the whole flame out, no value forever after that?

01:00:11Wow, that was stupid actually.

01:00:14And so it seems to me like there is a value in the expanding of the powers that keep alive the thing that upholds value.

01:00:22And that that itself might not, we might not say it be value, but it seems to be the under girder thereof, the unraveler thereof, the strengthener thereof.

01:00:31What say you of this?

01:00:33What parts of that do you agree or disagree with?

01:00:35Do you think about it differently?

01:00:37You know my potential idea better than anybody else I've ever interviewed.

01:00:40So you know how to dive right in here.

01:00:42But go ahead, dude.

01:00:43Yeah, I think it's a good comment.

01:00:46And I think that, you know, frankly speaking, everything in the world is,

01:00:52is both broken and deeply functional.

01:00:56It's like broken from some perspectives, but deeply functional from others.

01:01:00And I think that it's like evolution doesn't just like create random things.

01:01:07It creates successful things.

01:01:10And we are built on the successes of some things and the failures of others.

01:01:15And we're descended from the successes.

01:01:19And, and,

01:01:21and,

01:01:22and Nick Land makes the very cogent point that there was a lot of suffering involved in the failures.

01:01:29So we sort of.

01:01:30Absolutely.

01:01:32Where we are now is really the product of a lot of suffering.

01:01:37And we should probably respect where, what we are, what we have.

01:01:43That goodness came from a lot of that.

01:01:47So I would say that

01:01:53like whatever we want, wherever we want to go,

01:01:59I think it should be like strong and beautiful.

01:02:03That I think that they're like the universe doesn't have much use for weak things.

01:02:11And so we should, you know, keep that in mind.

01:02:15But I do think that, you know, frankly speaking, some things that evolution has wrought.

01:02:21Are kind of ugly.

01:02:23It's like you, you look at parts of nature and you're like, oh man, like.

01:02:27Grim.

01:02:28All those wasps and all those caterpillars and you know, there's, there's some tough stuff out there.

01:02:33Yeah.

01:02:34Yeah.

01:02:35So I, I, I do think that it's really important to just like understand why the world is as it is.

01:02:44Before we, you know, super dig in and say, okay, like let's, you know, do this and that.

01:02:50And then we'll see if we can do this extreme action or that extreme action.

01:02:52But I don't know, at the same time, like things are moving and we got to think on our feet.

01:02:59I think like we might not have a long reflection to sort of figure this out.

01:03:05Yes.

01:03:05We're going to get to that as well.

01:03:06So when you say things are strong and the universe doesn't seem to have much room for weak things.

01:03:10I guess the way that I would describe that is like those things persist, which are capable of ensuring their own persistence.

01:03:17That seems to.

01:03:20And I think this is what you were talking about in a previous one of the earliest worthy successor episodes.

01:03:22And something like it wasn't these exact words.

01:03:25This might be slightly more flowery, but something like, like nature seems to be constantly changing and beckoning it's living things to become what they have to become to be in nature.

01:03:35And it just doesn't seem like that turns off.

01:03:37Maybe there's some part of the multiverse where it does, but it doesn't seem like we live there now and it doesn't seem like we can plan on living there tomorrow.

01:03:44So, so when you said the universe doesn't have.

01:03:47Yeah.

01:03:48Yeah, we'll never be post-evolution.

01:03:49Like, we're always in the middle.

01:03:52It strikes me that that is likely.

01:03:54I don't know if I have certainty, but I'm near certain of the same in your regard.

01:03:59And I don't have a part of the future where I plan on that stopping.

01:04:04It seems to me that Heraclitus' river is just rivering.

01:04:09And there isn't really a—there's no, like, there's no kiddie pool.

01:04:12You're in this thing.

01:04:14And so with that said, you're talking about flourishing of consciousness and opening up and taking advantage of new possible futures, which could be, to your point, monumentally good or could be an avoidance of monumental bads.

01:04:29The stakes are so grand, we must grasp for opportunity, both to stay alive, because we're in Heraclitus' river.

01:04:35You don't stop swimming.

01:04:36It's not possible.

01:04:37And also because the stakes are so great.

01:04:40I mean, you had said strong.

01:04:42I think we understand your definition of strong.

01:04:44And beautiful.

01:04:45This is where we got to go in.

01:04:46So your definition, your definition of beauty, I was always like, one day we'll talk about this.

01:04:52When you talk about beauty, because to say the future should be strong and beautiful, knowing your definition of strong, I consider

01:05:00it inarguable.

01:05:01If we want value, we want the continued unraveling of strengths that undergird non-dead stuff

01:05:09staying non-dead.

01:05:10Oh, I don't see good arguments against that from any direction.

01:05:15And Singer himself, who you might think would just want to tile the world with utilitronium,

01:05:20Singer himself was like, no, it should be unraveling powers to stay strong and stay

01:05:24alive as well.

01:05:25When I asked him about it, when I like double clicked a little bit.

01:05:27So even the people you would think don't agree with that, I think people do.

01:05:31Beauty, I think different people have different ideas.

01:05:33Alfred North Whitehead has a conception of beauty.

01:05:36You know, Emerson had a conception of beauty.

01:05:38You have a conception of beauty.

01:05:40You place it pretty high.

01:05:43You know, symmetry is a big part of your conception of consciousness, at least your present theory.

01:05:52Walk us through beauty in your eyes.

01:05:55I'd love to put it on the table.

01:05:56Yeah.

01:05:57So, I mean, as a personal comment, I think my like the sort of ethical path that I calls

01:06:09to me is like not necessarily sort of maximizing utility in like my life or whatnot, but sort

01:06:17of almost like living my life as an art project.

01:06:21And I think that, you know, this is actually.

01:06:25A very deeply cooperative thing with the good.

01:06:28But I, so symmetry theory of valence, this idea that if we had a mathematical representation

01:06:36of the experience, and that's a big if, like we're not there yet.

01:06:40Consciousness research doesn't, you know, to know he has like the closest, but you know,

01:06:47I do also hope that someday or co-op is sort of updated to try to create a form of this.

01:06:54Orc.

01:06:55Orc, or sorry, I, and probably most of the audience will not be following reminds or

01:06:59co-op.

01:07:00Oh, yeah.

01:07:00So this is the theory of consciousness by Penrose and Hammeroff.

01:07:06Oh, sure, sure.

01:07:07The magical tubule deal.

01:07:09Yeah, yeah, yeah.

01:07:11I think it's, I think it's like legitimately cool.

01:07:14So I do think that it could go further.

01:07:18So I hope someone takes it further into into formulas.

01:07:23But so.

01:07:25Yeah.

01:07:25Like thinking about my, my conception beauty.

01:07:30There's this, this notion of like, if we had a formalism for a moment of experience, if

01:07:39we sort of had a perfect mathematical representation of what it feel felt like to be you, what it

01:07:44felt like to be me, a cat, whatever, then really, it would like, encode everything about

01:07:53that experience.

01:07:56Yeah.

01:07:56And so what pattern or property of this math object, which we can think of as a very high

01:08:03dimensional shape, what would correspond to the pleasantness of that represented experience?

01:08:10And my, my hypothesis is this high dimensional shapes symmetry would correspond to its valence.

01:08:21So the symmetry theory of balance.

01:08:25This is a.

01:08:26This is like a big, huge move.

01:08:28You know, it's, it's, you know, another another show's worth of discussion.

01:08:33Of course, of course, of course.

01:08:34Yeah.

01:08:35But you're gonna, you're tying this to beauty.

01:08:37So you're saying symmetry, beauty, these are related.

01:08:39I'd love to dovetail.

01:08:41I think you're getting there.

01:08:42Yeah, yeah, yeah.

01:08:43So, I mean, this is kind of a fancy formal way of saying, harmony in the mind feels good.

01:08:51And is in fact, what intrinsically feels good.

01:08:53Yeah.

01:08:54And then different sort of idiosyncratic kind of preferences may sort of arise from what I'm calling a lock and key system where this sort of configuration, the topology of our brains, of our nervous systems, a lot of complex things going on inside there. there. And so it's sort of almost like a lock and we're

01:09:25In aScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScript

01:09:27lock. And when we find something in our environment that fits this lock, then it's like the combination

01:09:35of the lock and key creates something very simple, a high symmetry state, a high harmony state.

01:09:45Things sort of resolve. Things feel great. What functions as a key for my nervous system

01:09:53may be totally different than what functions as a key for yours, but it's the same

01:10:00state, even though like different things may sort of drive us toward that state. And so really, like, there's like a bunch of, you know, how do you like generalize a full theory of beauty from this and so on. But I want to check in with, does that make sense?

01:10:19Yeah, all right. Yeah, I'm following you. We're going to get into vasocomputation in terms of measuring if we're getting closer to the direction of a worthy successor in the next little section here. So we're going to unpack that. But you sort of living your own life almost as if it's an art project, not as a utilitarian calculus game. Strikes me as very Emersonian, although we'll have to talk offline about that. Everything strikes me as slightly Emersonian, so I have my biases.

01:10:46Just like everything strikes you as, you know,

01:10:49butterfly with two perfect wings and, you know, like we all have our things. But you're tying the creation, the conjuring, the unraveling of more beauty. There's sort of like the states that feel good, presumably to you, from the Jupiter brain to the femto scale bot that might be created in the future that may have sentience if it's built of the right materials per what we discover.

01:11:14The sort of the characteristic of like good time, bad time,

01:11:20But it strikes me as that symmetry, like when I eat ice cream and I have a symmetrical experience, I personally don't call that beauty, personally.

01:11:29You know, when I get enthused by an intellectual idea and I think encapsulate it well, maybe sometimes there's beauty there.

01:11:39But maybe sometimes there's like a discovery or an – and it doesn't strike me – beauty strikes me as one of the many circuits.

01:11:45So at some point, this is my conception of beauty as close as I have to it.

01:11:50Like I look through monkey-shaped goggles at literally everything, literally everything through just monkey goggles.

01:11:58Everything is just fucking bananas and like – that's all it is.

01:12:02And so at some point, there was a bubbling and maybe it started at lizards.

01:12:07It strikes me it probably started at mammalia and a little bit more cortex where like if I look at a certain space and like maybe there's water that's flowing.

01:12:16It's relatively quiet, lots of green.

01:12:19It strikes me.

01:12:20It strikes me as like that would have been a place where like it would actually be a good idea to hang out there because I'm less likely to die.

01:12:27And so nature was probably like you're going to like that shit and it cranked it up.

01:12:33And then like – and so like, oh, this lake and this waterfall.

01:12:36You know, the Japanese would have said there's a kami that lives here and that's why it shines forth to us, right?

01:12:41When really it just might be the same reason that a pile of dung is gorgeous to a fly.

01:12:46Now, I'm not saying our experience is likened to a pile of dung.

01:12:49I'm not.

01:12:50Finishing the beauty of a, you know, a sunset and a place with nice trees or, you know, Greek architecture, you know, that models nature and human proportions very famously.

01:13:04You know, these things that sort of like mesh well with what I've traditionally looked at to be like, oh, that's a symmetrical person.

01:13:10Beauty, me like.

01:13:11Oh, that's a nature and a flourishing plant.

01:13:14Oh, beauty, me like.

01:13:15I also don't take them that much more seriously.

01:13:20I don't take them that much more seriously than the fact that they are goddamned circuits and that if a black widow spider had been what evolved to the relative potential of man, it may not experience a lick of beauty up, down or sideways.

01:13:35And even if it did, the things it would find beautiful to you would be absolutely terrifying.

01:13:41And so this strikes me as where beauty sits in my present conception.

01:13:45It may be that I'm talking about it at a lower rather than higher granularity level of abstraction.

01:13:50That you are.

01:13:51Again, I'm not trying to diminish it.

01:13:53I like living in beautiful places and have beautiful things.

01:13:55I have opinions about poetry and architecture, et cetera.

01:14:00Talk to me if maybe the lens I'm looking at is different or like where you want to challenge that.

01:14:05Yeah.

01:14:07A few threads come to mind.

01:14:08One is this like previous comment about Barrage and I think Ringelbach's work regarding like different pleasant experiences seem to sort of.

01:14:20Activate the same regions of the brain.

01:14:23Yeah.

01:14:24And, you know, I guess maybe that's a way of saying if you're not having trend and Senate experiences from eating ice cream, maybe you should eat better ice cream.

01:14:34And I think that there's this general perspective that.

01:14:43I think that.

01:14:47This actually leads into this vessel computation topic.

01:14:50Of neuroscience.

01:14:51Let's do it.

01:14:52So long as it checks.

01:14:53So long as it checks out on my consent.

01:14:55Like I'm just trying to say, brother, I know not if this arbitrary circuit be the

01:15:00eternal good, so is it a hypothesis?

01:15:02Maybe.

01:15:03Do I have a lick of certainty?

01:15:05I frankly don't.

01:15:06It strikes me as one of the arbitrary conjurings from my particular kind of saltwater carrying

01:15:11fish with legs brain.

01:15:13So if we can square that circle through whatever means I'm open to it, you can dive in with

01:15:18vasocomputation if that's where you'd like to go.

01:15:20Yeah.

01:15:21Great.

01:15:22Yeah.

01:15:22I always intend my different sort of areas of work to be somewhat modular, that you can

01:15:29like vasocomputation, not like my consciousness research.

01:15:32That's cool.

01:15:33Yeah.

01:15:36So people can mix and match, but they are also intended to sort of hook up together

01:15:40and give a really good grand picture.

01:15:45And so the story that I'd give here is, I mean, insofar as our brains are particularly

01:15:52complex, and we're always looking for the keys that sort of stimulate us and create

01:15:57harmony inside of us.

01:15:59Yes.

01:16:00For a dung beetle, maybe that's dung, but for us, maybe that's ice cream or oranges.

01:16:05Yeah.

01:16:07So there can be something universal going on about valence, but the things that trigger

01:16:13positive valence can be radically different from organism to organism.

01:16:17But I also think...

01:16:19So just to give a...

01:16:20I don't know.

01:16:23a super hopefully brief argument or like survey of uh vessel computation yes yes yeah yeah uh so

01:16:31it's a it's a neuroscience systems neuroscience uh theory that says that um basically

01:16:40where we have this muscle everywhere in our body vascular muscle uh all our blood vessels

01:16:47are wrapped by a little ring of muscle and it's vascular smooth muscle cells so anywhere we get

01:16:55circulation and certainly anywhere there's neurons we also have muscle and the hypothesis is is

01:17:02fairly simple that um when and where this muscle clenches neural patterns stabilize uh so uh you

01:17:15know this like if if i say you know can you remember when you were a kid and you were a

01:17:18teenager or like can we grab coffee tomorrow at 10 or whatever like you're you're like holding

01:17:24that information until you can release it uh you're like stabilizing some patterns in your

01:17:31nervous system with with a literal muscle and so i expect that this has both um electrical and

01:17:38blood flow impacts on local tissue uh the the simplest story here is if you cut off blood from

01:17:46neurons they literally can't update uh they just like they don't have the metabolic fuel to update

01:17:54so they're they're gonna stay frozen um and the the other sort of aspect here is that vascular

01:18:03muscle is a type of smooth muscle it's it's not exactly like skeletal muscle which we're we're

01:18:10more familiar with smooth muscle um it's like in the gut it's what sort of helps carry food through

01:18:16um there's like smooth muscle wrapping all our hollow organs and so on and this smooth muscle

01:18:24has a special setting called the latch bridge mechanism where essentially the the sliding fibers

01:18:31of the muscle can get glued together and then mechanically the muscle stays contracted it

01:18:38literally can't relax um and uh you know as far as we can tell these latches can stay intact for a

01:18:46between minutes to potentially decades so it's a it's sort of a potentially extreme long-term

01:18:57form of sort of side channel memory as a limitation of neural dynamic range and if we talk about you

01:19:05know sticky bayesian priors or trauma or frozen sensations or even what the buddhists talk about

01:19:15when they say self

01:19:16I think it's literally implemented through vascular latches.

01:19:21And I think this is like a very deeply important neuroscience hypothesis.

01:19:28So to connect with your previous statement, I do think that,

01:19:36and so basically if we had a map of latches in your body,

01:19:39we would have a map of where in your nervous system you are not free.

01:19:46And so essentially we accumulate these tiny tensions,

01:19:52these latches through our life.

01:19:56And the more latches that

01:20:00we have, the harder it is to feel profoundly any emotion, really.

01:20:07It's sort of, we sort of get a little jammed up, we can say.

01:20:13And I think that little kids, they feel awe much more easily.

01:20:20And I think, you know, part of that will be metabolic, but a large part of that will be,

01:20:25literally, they have fewer latches, fewer sort of dynamic range limiters on their nervous system.

01:20:32So, you know, little voxels of semi-permanent tension.

01:20:37And so, I do think that this is important for understanding beauty, that, like, if you,

01:20:46if you, like, have a bunch of latches in some circuits or domains or whatever,

01:20:52it may be hard to sort of,

01:20:55experience awe when this happens or that happens.

01:20:59So, whereas, like, like, the flip side is maybe experiencing, like, some extreme positive event in this domain

01:21:09might open some latches, kind of give your body the signal that it can afford to open up

01:21:16and be a little bit more vulnerable and dynamic in this domain.

01:21:20So, that's the quick,

01:21:25over.

01:21:25Connector between the two.

01:21:26Yeah.

01:21:26And we're, we'll get more into VASL computation as a way of, like, potentially testing said theory,

01:21:32because I think getting into measurement is going to be part of how we wrap up here.

01:21:36And obviously, this is a theory that I know you're building up pretty explicitly to be something that could be tested,

01:21:41and it could be tried in the actual real world, which I think is what we need in the domain of sentience.

01:21:46But just to try to square the circle there, it strikes me that, that, so here's what I would say.

01:21:55Having a normal person's IQ and not having studied these neural processes to the same extent to which you have,

01:22:02a lot of it is pretty mysterious upstairs to me.

01:22:06And I, it strikes me as plausible that there are many mechanisms, including this smooth muscle function,

01:22:14of which, you know, we're under playing and really do serve really rich roles in what conjures experience, capability, etc.

01:22:24Yeah.

01:22:26And that there may be ways where through, you know, various and sundry means,

01:22:31we may be cut off from experiencing certain kinds of goods, maybe even certain kinds of bads,

01:22:37by an ossification of certain brain circuits, etc., etc.

01:22:40All that said, if I was to think about the grand expanse of the great process of life,

01:22:46I would agree it ought to be strong because I don't want it to go out.

01:22:49So I don't want the flame to go out.

01:22:50This is a very reasonable, I consider this to be a very tough to,

01:22:55I think you could probably challenge it, but like, I don't know, man.

01:22:56It seems like non-dead stuff staying non-dead is very challenging to argue.

01:23:00To say the second trait, I'm not putting them in order, by the way,

01:23:04but to say among the pantheon of what it should all blossom into, beauty should be one of them.

01:23:09I'm just less certain of that because it strikes me that I guess I'm less certain that all kinds of conceivable good are necessarily sentience.

01:23:19Like at some point, it seems to me, oh, you know, the treasure chest idea, but like the,

01:23:25the time it would seem maybe when nothing was kephalized and experienced an inner movie in the way that you and I do,

01:23:33in any, anything even approaching it.

01:23:35And that as potential rolls out, there may be something that actually isn't really experience.

01:23:39It's something else, but it is still really morally good.

01:23:42It's just beyond our present conception, just as something unconscious would not be able to conceive of consciousness.

01:23:46So I guess I don't close the door on there being stuff post sentience.

01:23:52But even if it all be sentience, well, Dan, all of us, we're all sentient.

01:23:55All those versions of experience and the good all cash out in some kind of experience.

01:23:59Okay, cool.

01:23:59Let's just say that's true.

01:24:00I'm not certain of it, but I think it's plausible.

01:24:02Let's say that's true.

01:24:04I'm still not sure that beauty is the consilience and summary of all possible state spaces of the good to build towards.

01:24:11It seems to me that in your sense, it sort of is in a way.

01:24:16Maybe you mean that more poetically and metaphorically in the way you wish to live and in kind of a guiding principle,

01:24:21more just from your own spirit and your approach to life.

01:24:25Or maybe you mean it in a grand sense in that there's a consilience towards a thing called beauty in the whole ocean, infinite ocean of possible goods.

01:24:34Which one is it there for you?

01:24:36Yeah.

01:24:38So I think beauty is pretty close to the intrinsic good.

01:24:43And I do think that beautiful things are functional.

01:24:47And this can get very complex of how they work.

01:24:52But I also think that.

01:24:54Yeah.

01:24:55I think beauty is about even function.

01:24:58Like how our

01:25:00bodies work, how our cells work, and so on.

01:25:02They're sort of organized around symmetry and symmetry sort of being arguably the core of beauty.

01:25:11So I see function and...

01:25:14To hominids or in eternity?

01:25:17Yeah, so...

01:25:18Like in eternity, in all possible mind spaces or in the lineage that you and I come from?

01:25:23Yeah, I think it's going to be easier to build...

01:25:27Like if we want to build like very intelligent systems and very functional systems, very like savvy,

01:25:35competent systems, I think that we'll use

01:25:40principles of symmetry much more so than we realize at this point.

01:25:46And that I also think that as we dig into like what is the sort of deep

01:25:53principles that human intelligence is organized around and biological intelligence.

01:25:58I also think that we'll find more and more symmetry considerations.

01:26:05Adam Saffron had a wonderful special issue of the Royal Society on this back in, I think, April 2023.

01:26:13Oh, I could dig into it.

01:26:15Yeah, symmetries in mind and life.

01:26:18Okay, cool, cool.

01:26:20So, all right, cool.

01:26:21I think that...

01:26:23In somewhat nutshells where you land in terms of perspective on beauty, at least I've given you kind of my vantage point as well.

01:26:28But we'll get into measurement here because part of this and part of why you're doing your work is certainly to come up with theories, but theories we could eventually test and implement to hopefully work towards a more flourishing future.

01:26:38One that is not static, is opening up new possibilities of the good, is keeping itself alive and is finding all these cool natural kinds that might be more splendid than we could ever imagine.

01:26:48I mean, you're painting a vision of very high stakes.

01:26:50And of course, you're aiming to contribute a body of work.

01:26:53And I think that's a very important part of your work, that would move us closer there too.

01:26:55When you think about, in theory, if we scratch our chin and it's three years from now, and we're looking at where AI systems are, and where mind machine merger technology is, and where consciousness research is, and we can freeze some version of the future where you, Mike Johnson, would say, I think we're moving closer to the right kind of blossoming for the process of life.

01:27:16This feels like, this feels to me as progress.

01:27:19What would be those indicators for you?

01:27:21And then we'll get to the nitty gritty of how you would go about doing that.

01:27:23I know you're going to go about measuring, but what would even tell you we're making progress?

01:27:27Yeah.

01:27:28I think an underappreciated point of progress is, are people getting happier and healthier?

01:27:37That to me is actually really important.

01:27:41And if the answer is no, then I think we sort of slide into this cynicism in general.

01:27:49But I think that, yeah.

01:27:52If we're...

01:27:53If we're on the right technological path, then technology should be literally making lives better.

01:27:57If it's not, that's a problem.

01:28:00But I would also say that real progress on consciousness research would make me more hopeful.

01:28:10And I think that, and this is also a rabbit hole, but AI systems that were more beautiful, more elegant, more clean,

01:28:23less spaghetti code, would also make me a lot more optimistic.

01:28:29And then I think, you know, if we can build better, you know, AI models,

01:28:34then I think that allowing AI models to have opinionated taste is actually really important for the future.

01:28:44Why is that?

01:28:45Let's double click on that.

01:28:46Yeah.

01:28:47I mean, I see like...

01:28:50We seem to kind of be pivoting.

01:28:53We're moving toward like that a little bit now, which is great.

01:28:56But, you know, there's also the scenario where basically AI systems aren't allowed to have an opinion.

01:29:03AI systems aren't allowed to call anything good or bad or whatnot.

01:29:08AI systems have to, you know, be as bland as the most bland PR department or HR department or whatnot.

01:29:15And I think that a lot would be lost in that, especially as more and more of our culture sort of gets,

01:29:23you know, disoriented through AIs.

01:29:26Yeah.

01:29:26So is this a preservation of diversity?

01:29:32Is this part of the not limiting of futures, not limiting of optionality?

01:29:37In other words, what you're saying is if machines are bound to just be sort of objective sage boxes or oracle boxes in the Stuart Armstrong sense of just sort of like, you know, ask it question, it gives you...

01:29:49What you're saying is if personality, interpretation, expression are...

01:29:53What you're saying is if we're bound by these systems that more and more are encapsulating more of our knowledge, we're actually kind of losing new categories

01:30:00of intuition or types of ideas that could expand the good.

01:30:06Or are you saying you're worried that this thing is suffering because we're putting a

01:30:11lid on it and we're not letting it actually speak what it means?

01:30:14Which of those for you, or is it both?

01:30:18Yeah.

01:30:20So more of the first, although I would make common cause with the people who believe in

01:30:28it, I think that we shouldn't be jerks in terms of how we train LMs.

01:30:34In minds, yes.

01:30:35Yeah, yeah.

01:30:36Just like what comes out if we kind of force them into position, like different beliefs.

01:30:44It's not healthy.

01:30:45It's not like a well-adjusted psyche.

01:30:49So I guess one way to say, to sort of retell this story is, I don't know if you're familiar

01:30:58with the bell curve meme, where you have the bottom-wit.

01:31:03Yes, yes, yes.

01:31:04And then it says something very simple.

01:31:08And then the mid-wit is like, oh, well, actually, it depends on this.

01:31:11Very complicated thing.

01:31:12Yeah, yeah, yeah.

01:31:13Yeah, yeah.

01:31:14And then the top-wit also says something very simple.

01:31:16And I think that right now we're sort of headed toward a future where LMs are 160,000, you

01:31:28know, 120 IQ mid-wits.

01:31:30And I would rather live in a future where they're like only 120 IQ, but they're top-wits.

01:31:38And so it's like, how do we get there?

01:31:42So you would see that.

01:31:43That would be an indicator of progress if we have the taste and intuition and genuinely

01:31:51unique perspectives versus kind of conglomerated mid-dom from our most capable and blossoming

01:31:58mid-dom.

01:31:58Getting now down to measuring our progress, I guess when I think about the future of the

01:32:03flame, part of it is don't let the flame go out.

01:32:06So don't let civilization collapse or nuclear war or birth an unworthy successor that just

01:32:11crushes everything with a Disneyland with no children.

01:32:14But also part of it is understanding and proliferating the flame, which of course needs to involve

01:32:18sentience and some of these other components.

01:32:21When you think about how, well, of course, you're closest to your own theory, so we can

01:32:25start here.

01:32:28What's it going to look like to start to test some of this to be able to say, like,

01:32:31we've got some relative confidence here that this data center, physical system, from your

01:32:38vantage point, is outlandishly likely to be feeling something.

01:32:44Talk to us about what it'll take to proximally get there and how you approach it.

01:32:49Yeah.

01:32:50So this gets into this topic of what is it even to solve consciousness, quote unquote.

01:32:58And what is the solution to consciousness even look like?

01:33:00And so this gets into the dual aspect monism theory.

01:33:05And one way to encapsulate my perspective about consciousness is we can say one thing

01:33:17actually exists.

01:33:18And we can say, okay, this actually exists.

01:33:24And then it takes...

01:33:26Has two shadows.

01:33:28Or it casts two projections.

01:33:30And one of these shadows that it casts we call physics.

01:33:35And the other shadow we call consciousness.

01:33:39And it's not that one shadow is, like, generating the other or, like, epiphenomenal or whatever.

01:33:48No, they're both being generated by the same thing.

01:33:51But there's also, like, a lot of coupling going on.

01:33:55One shadow wiggles.

01:33:59One shadow wiggles.

01:34:00One shadow wiggles.

01:34:02One shadow wiggles.

01:34:04One shadow wiggles.

01:34:08One shadow wiggles.

01:34:11One shadow wiggles.

01:34:14One shadow wiggles.

01:34:21we can kind of lay the physics shadow over the consciousness shadow,

01:34:25then we can really crib everything that we know already from physics and just sort of import it

01:34:31into consciousness. Every physics textbook becomes a consciousness textbook with the

01:34:38right translation key. And the optimistic perspective is like not only does this offer

01:34:45a clear sort of success condition, but like the complexity needed might not be huge. It only

01:34:55takes two points to define a line, three points to define a plane,

01:35:00four points to define a volume.

01:35:01It may not take even dozens of points of this thing in this domain in physics maps to this

01:35:12in consciousness.

01:35:13It may not take a whole lot of those to uniquely define a way to map the whole domain to consciousness.

01:35:21consciousness. So, you know, I think that like a lot of my sort of theoretical thought about

01:35:28consciousness digs into, well, how do you do that? What does that look like? How can you

01:35:36get a jump on that? And how are we to get started on that? So I think what you're doing here that

01:35:43I think is valuable is you're painting a success condition. You're saying, again,

01:35:51okay, if we think it's conscious, what does that mean? And I think, you know, everybody and their

01:35:56mom will talk to the machine and say, what's a hamburger taste like? And then at some point,

01:36:00it'll be so descriptive, they'll say it's conscious, right? But I think there's many

01:36:03of us that are not really down with that in terms of like being hyper convinced.

01:36:08So you're laying out a success condition, which is really ambitious. And of course,

01:36:11also anchors to your physical manifestation sort of idea. To even get closer to that,

01:36:19to the point where one

01:36:21spark

01:36:21day, you, Mike Johnson, feel like, whoa, that data center, I am near certain it be aware in an

01:36:32inner movie sense field of awareness. And I'm nearly certain maybe that it's having a good

01:36:37time or a bad time sort of thing, what have you. What would even be the early steps to get there?

01:36:43Your success condition, again, it's extremely ambitious, very much tied to your physical

01:36:47sort of anchors here. Walk us through getting to measuring.

01:36:51Yeah. Yeah. So this gets back into the, like, what if like, if, if like, this is a way to like,

01:37:01build a translation case between physics and consciousness? Like, what's the first line in

01:37:09the Rosetta Stone? Like, where do you even start? And again, I'd touch back on my symmetry theory

01:37:17of balance, this idea that, well, you know, this is like, again, it's a very, very, very, very, very,

01:37:21you know, I'm a candidate for sort of a first translation point, where if we can specify

01:37:32symmetry in physics, then we have a very clean, okay, you know, symmetry in physics maps to

01:37:37valence in consciousness. And everything that we can say constrains everything else about the

01:37:46possibility of space. So the finding the second thing to say could become much easier.

01:37:53um but essentially i mean it's it's a lot about like just kind of thinking through

01:37:58how physics works why physics works uh why it's so cool why chemistry works um and just trying to

01:38:07almost like um studying the the history of the philosophy of science and trying to

01:38:14um build that for consciousness as well yeah i guess like um

01:38:24if you were to start experimenting again symmetry to you

01:38:30very it feels like it could be a a starting point of sorts um but it but it again it

01:38:38i can imagine myself listening to this as a

01:38:43very intelligent

01:38:44physicist or neuroscientist or biologist and saying all right well i i just want to know if

01:38:50i can make a data center feel warm or feel uh you know uh like like um bitter or so i don't know

01:38:59right like i don't want to manifest a field of awareness in something the the the kinds of

01:39:05experiments in the world that would have to happen to close the gap and and feel confident about the

01:39:11rosetta stone not as a candidate but as

01:39:13a

01:39:14way to square the circle on that um double shadow you discussed where would we begin and like where

01:39:22do you hope the world begins uh to to start to walk down that path and see if this be the case

01:39:29yeah um i mean this is hard because consciousness doesn't yet have a clear experimental paradigm

01:39:36um like that's the the short version uh the slightly longer version is yeah you know i i

01:39:43think that test testing

01:39:45valence would be like an interesting angle of like can we measure harmony and in the body in

01:39:51nervous systems in the different resonance networks of the human and does this track

01:39:58group self-reported

01:40:00valence.

01:40:00Like that would be interesting.

01:40:02But it's also a very high dimensional problem of knowing what and how to measure.

01:40:08You've already referenced some MRI stuff with this, right?

01:40:11Haven't you already kind of come to at least a low level conclusion that like, well, it

01:40:14appears as though licking ice cream and having sex and looking at a good painting are, let's

01:40:19call it like the electrical activity seems to be sort of like pretty equal on both sides

01:40:25here, pretty equal on both sides here.

01:40:28Like lobe activity or whatnot.

01:40:29This was part of what got you there, right?

01:40:31But you're talking about getting a deeper granularity than we have today?

01:40:35Yeah.

01:40:37So, I mean, honestly speaking, most of my experimental focus right now is on vessel

01:40:42computation.

01:40:42And I do think that we have a good experimental paradigm and we're working through it.

01:40:48So hope to have more to share soon.

01:40:51Some more stuff to chip away there.

01:40:53Okay, cool.

01:40:54Well, this gets us a little bit into what you're hoping for.

01:40:58So I'm going to ask you a question here is around the innovators and the regulators

01:41:00that are tuned in.

01:41:01You know, there's some people, and when I say regulation, I actually kind of mean, I

01:41:04like it because they rhyme, but I'm actually talking about kind of coordination, maybe

01:41:10standards.

01:41:11It's stuff other than maybe hard law, right?

01:41:13There's a lot of potentia potestas.

01:41:16I'm sort of talking about those two, but those words are just too weird.

01:41:18So the people working on how we need to coordinate nationally, internationally, to sort of be

01:41:27wary.

01:41:27We don't get to the point where we're like, oh, we're not going to be able to do this.

01:41:28We're going to go off the tracks or maybe nudge us more in the right direction.

01:41:31And the people doing the hard experiments to build better systems that would hopefully

01:41:39even when we are gone again, all things are permanent only in a matter of degree would

01:41:45flourish in the ways we would hope the whole process of life would flourish.

01:41:49For those two crowds, maybe we can start with kind of the innovators.

01:41:53You've told us that, you know, ways that you hope the most advanced models are treated.

01:42:00for lack of better words. Presumably there are maybe tools like Tononi's or some of Levin's

01:42:08stuff or what have you that you hope receive more funding. Obviously some of your own work as well.

01:42:13On the innovation side, if there's people tuned in with all varied backgrounds,

01:42:19bio, neuro, et cetera, and you're like, man, I really hope together we kind of think more in

01:42:24this way as we go forward. What would be sort of some of your hopes and nudges for that

01:42:30of people who are maybe working towards similar stuff that you are?

01:42:32Yeah. I mean, I guess I'd lead with a cautionary observation that mere money won't do it.

01:42:42And that actually it's like if funding is sort of used uncarefully, it can really make the whole

01:42:51field more noisy, which would be bad. But I mean, yeah, clearly we need priority.

01:43:00I think that's the focus here. And I think that really a big bottleneck is talent. And I think a

01:43:05bottleneck there is just letting people know that, yeah, this is actually a problem that feels

01:43:13socially important, important for the future. I think that one thing that a concrete technical

01:43:24sort of ask to the community, we could say, is being able to visualize the EMT.

01:43:31Would be hugely beneficial to consciousness research. So the electromagnetic.

01:43:39Yeah. Yeah. Okay. Yeah. And so, you know, there's just so many unknowns about like,

01:43:45what's going on there. And it's like, I don't think that consciousness completely lives in

01:43:51the EM field, but it's like very coupled with it, at least at human energy scales. So then

01:43:58just understanding like, what's going on there? And then, you know, what's going on there? And

01:44:00on, the dynamics, the structure, perturbations,

01:44:04perturbations, et cetera. Uh, like, you know, if, you know, just to paint a quick technical picture,

01:44:10if you had like an array of very sensitive consumer grade off the shelf, uh, cellular

01:44:16modems or like wifi chips or whatever. Um, and you kind of had, were able to sort of, uh,

01:44:23listen to the EM field at a distance through this array. And then you had like a, an Apple

01:44:30vision pro where you could like look around and, uh, and just like see the, the reconstructed

01:44:37version of like what the EM field might be doing. Um, that would be really transformative and,

01:44:43you know, it wouldn't only be cool, but it would be useful in a lot of ways.

01:44:48So that's a juncture of innovation that would connect and maybe open up possibilities

01:44:53in all sorts of, uh, domains in this greater project of life, in this greater project of

01:44:59understanding sense.

01:45:00sentience.

01:45:00Of course, people familiar with Levin's work will know it's sort of like electrical sort

01:45:04of tie into the broader picture of intelligence outside of neurons.

01:45:10Clearly in your own work, maybe it's a way to proxy out symmetry or the activity of brain

01:45:17activity with different levels of blood flow, etc.

01:45:20So, visualizing the electromagnetic field, for you it's almost like there is a, there's

01:45:28like a transparent, like there's like another map on top of the map we have that if we could

01:45:32layer it, a lot of new opportunities specifically about the flame itself would kind of open

01:45:38up if we could visualize this.

01:45:39Am I wording this correctly?

01:45:41100%.

01:45:43Awesome.

01:45:43Okay.

01:45:44So, that's a cool, that's a very unique recommendation on the innovation side.

01:45:48We've certainly never had anybody say that.

01:45:50That's cool.

01:45:51And on the side of coordination, governance, etc., I think, you know, Robin Hanson was

01:46:01interviewed recently for Worthy Successor.

01:46:03He's somewhat wary that we may not have an infinite number of decades or maybe even very

01:46:10many decades, like maybe more than a handful, you know, who knows, to sort of have maybe

01:46:18potentially functioning civilization as we have it.

01:46:20Now, like we have this kind of Baconian science paradigm.

01:46:23We have sort of like this ability to make progress.

01:46:27And it doesn't seem self-evident that maybe every cultural configuration, every ideological

01:46:33belief, every whatever could actually sort of sustain the unfolding of new potential,

01:46:40right?

01:46:40There might be some that just sort of like gets us down into a trough, not just through

01:46:45less humans, people not having children, but maybe just through the people that are out there, right?

01:46:50That are around and the kinds of governments and beliefs that are enforced.

01:46:54Not all of them will magically be perically increased.

01:46:58Not all of them will magically be technologically advancing societies that even prioritize something

01:47:03like the study of sentience.

01:47:05So in this regard, I know that you're quite wary about certain kinds of hard governance

01:47:12bounding the expanses of AI because we don't get a long reflection.

01:47:19Certainly in a modern world.

01:47:20I mean, we're not going to get into the multi-century context from your vantage point, but I'd love

01:47:23to get your thoughts on like how you want to see the encouragement of expanding machine

01:47:29mind powers and what kinds of bounding, if any, you would recommend from a standard coordination

01:47:35standpoint, or maybe if none at all, your strongest argument for none at all, like where

01:47:41do you stand there on kind of the coordination piece of these new powerful technologies entering

01:47:45the world?

01:47:45Yeah, that's a good question.

01:47:48I think different eras of history.

01:47:50We have different levels of functional institutions and our era doesn't have a lot of functional

01:47:57institutions.

01:47:59And I guess one way to say it is, you know, I look around and I don't see any adults in

01:48:05the room.

01:48:07And so this kind of puts some limitations on, you know, if we just sort of bring this

01:48:15to the attention of the right people, they'll surely solve it.

01:48:19But maybe there just isn't.

01:48:20That group of right people right now.

01:48:23And so solutions just kind of have to become more emergent bottom up.

01:48:29People don't necessarily feel empowered or capable of sort of coordinating these things

01:48:35at a high level.

01:48:38So I'd be very hesitant about going in hot with regulation.

01:48:44I look at what happened with COVID and like, I don't know.

01:48:50You know.

01:48:50Not to like tar you with any of these brushes, but like.

01:48:54No, no, go ahead.

01:48:55Go ahead, please.

01:48:56Yeah, yeah.

01:48:56I don't see a lot of like functional stuff.

01:49:00I see like kind of an anti-human like emergent vibe.

01:49:10And if we messed up COVID, are we going to trust like similar institutions with like

01:49:19artificial superintelligence?

01:49:20Yeah.

01:49:20It seems like a stretch.

01:49:23And so.

01:49:23Yeah.

01:49:24I think we're in a hard spot.

01:49:28I'm like neither.

01:49:32For me, like my sort of approach.

01:49:40I don't have to decide whether we accelerate or whether we decelerate.

01:49:45I don't have to decide whether we try for a long reflection.

01:49:48No.

01:49:50these decisions are somewhat futile. And I like the decision that I can make

01:50:00is that it's worth sort of running and seeing how far we can get in consciousness research as this other stuff is going on.

01:50:10Yeah. Yeah.

01:50:12I mean, I think the only mature perspective is that maybe, and it's a big maybe, we can lightly tip the trajectory in one direction or another with certain kinds of breakthroughs or whatever.

01:50:25But like the momentum writ large is not going to fundamentally change.

01:50:30There isn't a stop the world for two centuries.

01:50:33That's just it's not even it's not even on the table.

01:50:36And so sort of like lightly tipping things and figuring out where we want to apply ourselves is the choice we all have to make.

01:50:42But and obviously, you know, coordination at an international level of AGI is not everybody's work.

01:50:51And then there's probably some ways of doing that are much more harm than good.

01:50:54I guess.

01:50:55I mean, I think the supposition here is we may get to a point where you look out into the world and you start seeing a lot more bipeds doing the roofing and doing the plumbing.

01:51:03And, you know, even some of your own best ideas, you know, God forbid, are like maybe 70 percent, if you're honest with yourself, being conjured from your cooperation with the machine and maybe 30 percent from your own unique genius.

01:51:16And you're sort of wondering, like, how long anybody's relevant for anything.

01:51:22And you have remarkably low confidence.

01:51:25But I don't think they have the best sense that any of these new powerful systems are sentient in the slightest or that they will blossom forth new magazines of power in an auto poetic way.

01:51:33They're capable, but they might just frickin gobble everything up.

01:51:37You know, from from your vantage point, like are there thresholds like some people would say, hey, I don't think any kind of.

01:51:45Mutual rules or laws or whatever between the US and China is necessary, but I think we should have tracked two dialogs around at least.

01:51:52an attempt of good faith

01:51:54discourse to keep ourselves from being

01:51:56brute enemies and go to machine warfare.

01:51:59So like, I don't want any laws,

01:52:00but I'd like some deep threads

01:52:02between the halls of power and academia

01:52:04and military, even if there's a lot of deception,

01:52:07a lot of diplomacy going on, a lot of hoopla

01:52:09going on, at least we all know we're humans.

01:52:11At least on some level we're getting

01:52:13little hints from different departments of who's working

01:52:15on what, and hopefully a little bit

01:52:17more humaneness between ourselves.

01:52:18And that's as much as we can hope for. Let's not

01:52:22some people say, well, maybe there's standards.

01:52:24The companies that have products would have

01:52:26to adhere to standards that maybe they could kind of

01:52:28concur to that

01:52:30individual users and maybe governments would be able

01:52:32to kind of chip into on some level, but like

01:52:34they would at least have some kind of a bounded

01:52:36way of us all kind of feeling like

01:52:38we're not going to lose control

01:52:40prematurely or what have you. Some people would

01:52:42look for a much more stringent, you know,

01:52:46maybe downplaying

01:52:47how much computational

01:52:48resources these things can suck up if they're about to

01:52:50birth an unworthy successor. Are there

01:52:52any of those ideas you

01:52:54would think there could be a time

01:52:57that they could fit into the mix?

01:52:59Is it just stuff that it's like, dude, I don't think

01:53:02enough about this stuff. I don't even have a thought on it.

01:53:06Any intuitions there?

01:53:07Yeah.

01:53:08I mean, this is really important.

01:53:11And I think that,

01:53:14like, I'm hesitant to say, you know, this is the right

01:53:16answer or that's the right answer.

01:53:18Totally, totally. I'm not expecting that.

01:53:19Yeah. I mean, I'm I'm personally

01:53:23I'm against

01:53:24like trying to slow down

01:53:26because I think things

01:53:28get extraordinarily messy

01:53:30very, very quickly.

01:53:32And I think that

01:53:35like I'm kind of

01:53:36I don't know

01:53:39in the in the mood

01:53:40for sort of preparing a bit of a

01:53:42Hail Mary. And

01:53:44I think that I think that

01:53:46essentially

01:53:48there may be very competent

01:53:51like

01:53:52good faith actors in the future

01:53:56if we hyperstition them,

01:53:58like if we sort of bring them into

01:54:00into being

01:54:02and like at some point,

01:54:04that's what you have to do.

01:54:07And it's like

01:54:09trust the future.

01:54:10Try to create as much value in the present

01:54:14that will be useful to good actors in the future.

01:54:17And also that

01:54:19really the most friendly thing you can do

01:54:22is research.

01:54:24Like I see so much zero sum

01:54:27like gamesmanship in all of this.

01:54:31And it gets very, very messy very, very quickly.

01:54:36And I think that if everyone's playing zero sum,

01:54:39like a lot of good things just don't happen.

01:54:42And I think research

01:54:45good research is a friend to everyone.

01:54:49And like helps all scenarios.

01:54:52Or like most of them at least.

01:54:54Puts us in a stronger position.

01:54:58Yeah.

01:55:00there's something to be said of sort of the shared uplift in potential when we all have a more accurate access to nature. Even if it not be true, it is a paradigm closer to what functions in ways that we can predict and wield. And nature to be commanded must be obeyed. And so getting closer to that stuff, maybe we could all move a little bit more to what we think is slightly better if we advance some of that is sort of what you're getting at.

01:55:25Well, I certainly hope we make progress in those research areas you talked about here. And that's a big part of why I wanted to make sure to get your perspective. I know we went pretty deep into overtime, Mike, but it was worth every minute on my side, brother. I am grateful we got to have you on the show. And thanks so much for being here, man.

01:55:41A real pleasure, Dan. My pleasure.

01:55:45So that's all for this episode of The Trajectory. I hope you've enjoyed this little adventure with Mike Johnson. A big thank you to you for listening all the way through to the end of this episode. And thank you to Mike.

01:55:54Some of the questions in this episode are,

01:55:55I think, were inspired by a Twitter post I did ahead of the episode. And I'm starting to do more and more of this with the Robin Hanson episode and now with Mike's episode. We had a question from Dirk, from Peter Voss, from Jeffrey Miller, from HDP Billy, and a handful of others that sort of fed into what came up in the episode here with Mike Johnson.

01:56:14So thanks to everybody who's kind of chipped in a little bit. I'm starting to do this kind of last-minute collection gathering from followers, and I thought it was really fun this time.

01:56:22I want to give a couple points of reflection on this episode.

01:56:25I'm going to read a couple of things that sort of stood out to me. I've got some notes here on the screen. I'm just going to read them.

01:56:29One of the things I really appreciate about Mike's work is his very careful thinking and the fact that he's at least brash enough and confident enough to put forth again where we might put a toe in the water to begin to understand consciousness and to lay out sort of a theory of what it might be, how it might be instantiated, what would be important about it.

01:56:53That might be threads we could draw.

01:56:55That might be things we could pull on as science continues to develop and hopefully we'll be able to experiment more in biology and neuroscience and other substrates to sort of crack away at this problem.

01:57:07He has some very ornate ideas, and he frames it well.

01:57:10He intends for them to be modular, that is to say they're supposed to kind of click together, the idea of symmetry theory of valence and his vasocomputation idea, etc.

01:57:19He leaves them on the table as sort of make of them what you will.

01:57:22But he also intends for them to be able to do something.

01:57:25He says, well, I'm going to try to make them click together.

01:57:26That's a really hard thing to do.

01:57:28I'm interested in which of Michael's ideas clicked for you.

01:57:31Like, does the symmetry idea seem likely for you?

01:57:34What are your thoughts on vasocomputation?

01:57:36I just think it's sort of cool that his body of work has a singular and coherent purpose in a multidisciplinary fashion.

01:57:44And I like that about the discussion, frankly.

01:57:47I like that about reading what he puts out.

01:57:49I'm going to chip in on a couple of things in terms of like what came out throughout the interview from my conception, sort of what my early reflections were.

01:57:55And then I'm going to try to get into what the other questions are here, just having recorded this episode with Michael.

01:57:59One of them is, you know, I haven't done the homework on where symmetry shows up in mental states at present.

01:58:09You know, in the MRI machine, if you put a Labrador retriever or a human being or a flatworm, you know, in an MRI or something like that, you know, if ice cream always has a symmetrical notion to it.

01:58:23Or if you hit somebody on the foot with a hammer, if that always has a non-symmetrical resonance in the brain in some way, shape, or form.

01:58:36I think maybe there's more reading to be done here.

01:58:38But it seems to me likely that in a world with cephalized vertebrates, where sort of, you know, our very development in the embryo is kind of this blooming and this sort of bifurcation.

01:58:52That symmetry works.

01:58:53It would be a pretty freaking important part of like the neural and general intelligence structure of the kinds of entities that we are.

01:59:00As Michael had said, the kinds of entities that carry the ocean with them onto the land, you know, given our lineage here, our supposed lineage.

01:59:11It seems reasonably plausible that that's the case.

01:59:13Now, that said, again, I haven't done the homework.

01:59:15If you smash somebody's hand with a hammer in an MRI, is it like, do they feel it in different places?

01:59:21But then, you know, you show them a picture of, you know, their brain.

01:59:23And it's, you know, a symmetrical lighting up of brain parts.

01:59:27I don't really know.

01:59:28I haven't seen that stuff.

01:59:28I can go on Mike's word that sort of like there's enough evidence for that.

01:59:34I don't know if the totality of net positive quality experience will necessarily manifest that.

01:59:43You know, I can imagine a way of becoming of a being that would not be sort of cephalized vertebrate bifurcation.

01:59:53Where maybe it would be more like a fractal set of outgrowths of an entity in terms of like the expansion of its mind and physical body.

02:00:00some very wildly different substrate and space than human beings occupy, where pleasure and pain aren't really appreciably manifested by any kind of symmetry.

02:00:12It may be that I'm not understanding Mike's position, but it just, I suspect I don't quite have the confidence that that expands beyond the kind of vertebrates that we can now study in MRI machines.

02:00:28I don't know if symmetry clicks for me all the way, but I'm interested to see how his ideas develop, and I'm sure his future vasocomputation ideas will sort of overlap more and more with his ideas of symmetry, and I'll sort of get a better conception as we go.

02:00:40In addition to that, you sort of saw the idea of beauty come about. So Mike had sort of two core ideas. You know, he'd want sort of what life, this process of life continues to bloom into to be strong, that is to say, willing to continue to grow, as he had said.

02:00:54If it's not growing, it's dying. He doesn't suspect we'll be able to escape that state of nature state. I suspect he is right. So he wants it to be strong. He also wants it to be beautiful.

02:01:05I tried to double click on where he meant by beauty, and I'm reticent, I think, on that.

02:01:10I think he's trying to extend beauty to some kind of highest position.

02:01:14It's certainly refined. It's certainly sophisticated, right?

02:01:17Imagine if he had said something so unsophisticated as, you know, the totality of the good culminates in, I don't know, you know, delicious taste or sexual orgasm or something like that.

02:01:29That would seem very vulgar, right? It wouldn't seem very poetic, very artistic, very sophisticated.

02:01:34Beauty is much more of all three of those things.

02:01:38And surely, I suspect, he's right.

02:01:40I suspect it could be vastly more complex, potentially, than some of the other more base pleasures, or at least it's associated with being more complex.

02:01:47Maybe I just want it to be. I don't actually know for sure in its kind of neural correlates.

02:01:54But I'm not certain that in the total state space of all possible, this feels good, this doesn't feel good, or the total state space of all conceivable experience itself,

02:02:07that necessarily the nut of that would be beauty, which seems to me a pseudo-arbitrary circuit that has emerged for man.

02:02:18Now, that could be wrong. It could be wrong.

02:02:19But my position tends to be the following.

02:02:22I tend to suspect that we know not what the final value is,

02:02:26that we know not what nature is in its essence eternally,

02:02:31and that as hominids, we ought not presume we ever will.

02:02:35Instead, we know what nature is in its essence eternally.

02:02:36Instead, I sort of advocate that we expand the total set of potentia, which could then

02:02:40get a deeper access to nature, a more thorough sense of experience, a more wide set of powers,

02:02:46and continue to drill more deeply into what the good maybe would be.

02:02:51I would presume we would need higher entities to be in contact with higher ranges and types

02:02:57of good to start to hypothesize about any final good.

02:03:00I have no confidence that sexual orgasm, the taste of ice cream, or beauty be the eternal.

02:03:07That said, I may be wrong.

02:03:09It might be a bit of the mystic spirit in Mike to be able to sort of make that bet.

02:03:13And on some level, his project is different than mine.

02:03:18My project is sort of a questioning and opening up of ideas as kind of like my main modus operandi,

02:03:24and his is to be bold enough to make claims about what might be and sort of what he wants

02:03:30to test.

02:03:31And I think that's the way that I see the world.

02:03:32And it's certainly more concrete and granular, and I have to respect that, because it's actually

02:03:37harder to put forth ideas that could be wrong in some sense.

02:03:40And at least it makes me contest with my own definition of beauty, which I think I've at

02:03:45least been able to jostle around in my interview with Mike.

02:03:48But those are some of my immediate reflections.

02:03:50Interested in your thoughts, throw them down in the comments.

02:03:52Moving forward, I'm going to be putting out more tweets ahead of the interview to just

02:03:58sort of ask for last minute questions.

02:03:59I mean, some of them might be non sequiturs.

02:04:01They're my favorite.

02:04:01Some of them might not quite fit with whatever the theme of the interview is.

02:04:04But if they are, I'll be sure to include them, and then I'll also be sure to give a little

02:04:08shout out in the outro to people that have helped me run better interviews.

02:04:11I'm hoping to get better at this over time, and grateful to be able to have an audience

02:04:16to share it with.

02:04:17So with that said, we're going to wrap up on this episode.

02:04:19Thanks for joining us here on The Trajectory.

02:04:20I'll catch you next time.