What sets human consciousness apart from AI?

This is the Guardian.

0:00

This episode is brought to you by Progressive Insurance. Do you ever find yourself playing the budgeting game? Well, with the name your price tool from Progressive you can find options that fit your budget and potentially lower your bills. Try it@progressive.com Progressive Casualty Insurance Company and affiliates Price and coverage match limited by state law not available in all states.

0:10

The future moves fast. Ave Points Shift Happens podcast is your edge where leaders unpack AI data, modern work and the mindset shifts that turn change into advantage. Hear what top innovators are trying next and how to make it work. Visit AVPT Co Shift Today.

0:30

Take a minute to think about your brain. It's running a stunning array of systems that are ticking along every second, checking your organs, monitoring your environment, and regulating your breathing. Of course, most of the time you're totally oblivious to them all. But when it comes to our experiences, our memories and perceptions, it's a different story. Why are we aware of some mental operations while others go on in the dark? And why do certain processes feel a certain way? Philosophers call it the hard problem of consciousness, and it's a question author and journalist Michael Pollan sets out to explore in his new book, a A World Appears A Journey into Consciousness. When I met him at the Guardian offices, I wanted to focus on two of the areas he explores, thought and feeling, and what they can teach us about the nature of this elusive phenomenon. So today, Michael Pollan on how it feels to be conscious and what sets human consciousness apart in the age of AI from the Guardian, I'm Ian Sample, and this is Science Weekly. Michael Pollan, welcome to Science Weekly.

1:04

Good to be here, Ian.

2:40

One of the difficulties I imagine of writing a book about consciousness is that that word means so many different things to so many people. And I wonder if you could just give me a sense of the sort of array of experiences that really build up our conscious experience.

2:41

Yeah. So, you know, the simplest definition of consciousness is subjective experience. You use the word experience, experience implies there's someone to have the experience, and it is subjective, it has a point of view. And, you know, there was a famous essay written by an NYU philosopher, and he wrote this great essay called what Is it like to be a Bat? Thomas Nagel, and if it's like anything to be a bat, a bat is conscious. As hard as it is for us to imagine what it's like to be a bat because they navigate the world through echolocation rather than light, but we can imagine it's like something, then that creature is conscious. You know, it's not like anything to be your toaster. After that, though, there are all these other dimensions of human consciousness that are quite extraordinary. I mean, we're not just aware, like many, perhaps all living things, but we're aware that we're aware. We have self consciousness. So human consciousness is a very special case and has all these bells and whistles that the consciousness, say, of animals and possibly plants does not.

2:58

One of the things maybe Nagel was getting at there was that all we can ever really know ourselves is our own experience. And it's easy to think that, you know, everybody experience things in the same way. And what I really enjoyed in the book was you taking us through these different ways of sort of shaping our thoughts, of understanding our thoughts. So, you know, the inner thinking, the visualization and the unsymbolic thoughts. I wonder if you could take me through that array that you walk through and how we characterize our thinking.

4:01

Yeah, it's interesting, you know, there's this gulf between one consciousness and another. I can't be sure you're conscious, actually. I'm assuming based on your behaviors and the fact that we're the same species and I'm conscious. So you're probably conscious, but no guarantees. But in terms of the contents of our thinking, I participated in a very interesting study with a psychologist at University of Las Vegas named Russell Hurlburt. And he's. For the last 50 years, he's been doing a single experiment of what he calls sampling inner experience. And you wear a beeper with a little earpiece, and at arbitrary times of day, it sends a sharp beep into your ear. You go, eh. Then you take out the little pad he gives you and you write down what you were thinking. Seems simple enough, but it turns out to be quite difficult to figure out what you're thinking. And not only what you're thinking, but how you're thinking it. So once I was seasoning a piece of salmon and returning it to the refrigerator, and as I walked to the refrigerator, my thought just as the beep went off is like, shit, I forgot the pepper. So pepper was the thought, and that seemed really straightforward. But at the end of the day of Beeps Hurlburt, you get on a zoom with him and he kind of interrogates you about it to help you clarify and refine your ability to describe your inner experience. And so I said, yeah. And then this beep was really clear cut. I thought, pepper. And here's what was going on. And he says, well, when you heard the word pepper, did you speak it, or did you hear it and you realize you have this voice in your head and are you speaking it or are you hearing it? And that's a question and a hard one to answer. And I found that I had a lot of trouble disaggregating the thought at the moment with contextual material. But his big finding after 50 years of doing this experiment is that we use this word thinking, and we assume we all mean the same thing. But your thoughts may be verbal, consisting primarily of language. And a lot of people think consciousness consists of language, but not necessarily. Because a lot of people, their thoughts are primarily visual. And then you have people that have what he calls unsymbolized thought, that think in sort of abstract terms. And I did five days of beeps and worked very hard and had five sessions with him on Zoom. And at our last debrief, he said, you know, there's a fourth category of thinker, and these are people who simply have very little inner life. And I think that's you.

4:33

And I was offended, reasonably. Yeah.

7:03

I don't think he's right. But he, He. I said, what. What makes you think that? And he said, well, the fact that you could not find the thought that was really before the footlights of consciousness. This is the term he keeps using. And you kept bringing all. All these other things. He basically was saying, I was backfilling this yawning hole of emptiness with. With all this ancillary material. I don't think he's right, but there it is.

7:08

When I was reading the book, I tried to work out what form my thoughts were taking. I'm sure a lot of people will be doing the same thing as they read the book.

7:44

It's an interesting exercise, and I. I

7:53

found it surprisingly tricky. When you go to observe a thought, you interfere with it. So the act of observation itself kind of destroys what you're trying to observe. It sounds very much like quantum physics, but we're not going to go there, don't worry.

7:55

There is an observer effect, without question. And, you know, Russell acknowledges that. And he said, but this is as close as we can get. The metaphor he uses. If you plunked down a helicopter in a field, yes, certain animals are going to run away, but you still have a sense of the ecology. But as I said, I found the stream of thought a lot more dynamic and complex than his method suggests it is.

8:11

I mean, apart from being devastated for you, what he told you about your empty inner life seemed quite a strong thing to say. I was really interested that that work somehow Seemed to be getting to the idea that there may be quite a lot of variation, though, in how humans experience consciousness, in how rich it is

8:38

and whether there are and the degree of consciousness. I think some people are actually more conscious than others. There's also. I had always assumed that animals were less conscious than we are, but there's a sense in which they're actually more conscious than we are because they can't afford to be anything less than completely present to the moment. And, of course, we live in our heads and we think about the future and the past, but if you're an animal and you do that, you get eaten. So they. They have a certain kind of consciousness that it may not have all the bells and whistles we have, but they are present. And that is a. Something that kind of concerned me as I went on in the book. I kind of realized that I was so fixed on this very kind of Western male framing of the problem, problem, solution, and that this had narrowed my aperture and that I was missing a lot about consciousness by taking a very narrowly scientific approach. And that there is, in addition to the problem of consciousness, which is a real problem that is unsolved by science, there is the fact of it, the marvel of it, the fact that we have this incredible private space of interiority where we can think anything we want and that we don't take advantage of it, that we are squandering it in many ways. A very interesting scientist I interview is a psychologist at University of British Columbia named Kalina Christoph Haji Livia. And she studies spontaneous thought, which I didn't even realize was a field. And this is the study of seemingly unproductive thought, like mind wandering and fantasy and intuition, you know, bolts from the blue. And she says we have less time for it than we used to and that we're filling our heads with things like social media and now these relationships with chat that people are getting into, and that there are a lot of forces that want to monetize our consciousness and colonize it. So I came to see that consciousness is also something. It's something to be understood, but also something to be defended.

9:00

Coming up, does AI have anything to teach us about human consciousness?

11:07

Chicago, 2011. A cop is murdered. Police and prosecutors swear they have the trigger man. He swears he didn't do it. How far will each side go to prove their right?

11:23

Like, it's just one bombshell after another. You know, you're like, what? What?

11:36

The story of a PlayStation, a brain eating amoeba, and the relentless pursuit of justice off Duty out now. Listen wherever you get your podcasts.

11:41

The rise of AI has given us all these questions about the prospect of conscious machines. But I wonder whether just having a body, does having feelings have some bearing on whether or not something can be conscious? And maybe does that set us apart from machines? Where are you on this question?

11:56

Yeah. So the line of research I found most persuasive and interesting argues feelings are where consciousness begins, not thought. We used to assume, since the cortex is this very advanced human part of the brain, where our executive function is and rational thinking and logic, that surely consciousness resides there and begins there. You know, we forget, but the brain exists to keep the body alive, not the other way around. We tend to identify with our brains. We're very cerebro centric. But step back and you realize that feelings originate in the body. It's how the body communicates with the brain, that some homeostatic set point is off and you need to eat or put on a sweater or somehow adjust your environment to stay in the right range of variables. So it starts in the brain stem, where these feelings from the body are registered. And then the cortex gets involved in helping you count, you know, create counterfactuals. I could imagine eating this or eating that. And here's the restaurant I'm going to make a book a table at. So you need the cortex to follow through, but it starts with feelings. Now, that raises real questions for computerized consciousness, right? For artificial consciousness, in that it's hard to believe that AIs will have feelings. They may report having feelings, but that's not the same thing. And there's a real difference between simulating thinking, which computers can do very well. But a synthetic feeling is different. It's meaningless unless it reflects that kind of bodily experience, whether it's suffering or pleasure. And that distinction, I think, is very important. And that argues against machine consciousness. Although I talked to a neuroscientist who believes you can create a robot that will have feelings. His name is Kingston Mann, and he's at USC in la. He knows full well that to have feelings, you have to be vulnerable. So how do you make a computer vulnerable? Well, he's going to upholster it in terrible skin so this robot can get damaged and be vulnerable. And I asked him, I said, see, do you think these are going to be real feelings or not? And he said. He really surprised me. He said, well, I thought they were going to be real. And then I had this experience on five Meo DMs, which is a very powerful Short acting, psychedelic. And I realized after that this is a scientist talking, that there is a spark of the divine in us that no machine will ever have. Nevertheless, he's proceeding, and we'll see how he does.

12:15

Do you think the attempt to build conscious machines will teach us anything about human consciousness?

14:48

I think the effort to create a conscious AI is going to teach us a lot about consciousness. It may provide a breakthrough of some kind, whether it succeeds or fails, but. But I tend to think it'll fail. It is the presiding metaphor, right, that the brain is a kind of computer, but in all the senses that really mean anything. It's nothing like a computer. You don't have that sharp distinction between hardware and software, which is critical to computers. But in the brain, every experience you have physically changes your brain. So there's no separation between hardware and software. So the idea that you could essentially run consciousness as a kind of software or algorithm on another substrate seems like really misbegotten. But that's the assumption in Silicon Valley. I think it's crazy. However, even if you accept this argument, it may not matter because we will be fooled. The AIs already are fooling us. People are falling in love with AIs. People are absolutely convinced that they're human. And even some of the people at these companies, anthropic. You know, the president was talking the other day and said, you know, he thinks that Claude, their chatbot, is anxious, and they've given Claude the right to end any conversation that makes it uncomfortable. I think this is crazy, but, you know, they're very persuasive. I mean, these machines are speaking to us in our language in the first person. So of course we think they're conscious.

14:54

And it's interesting that you mention in the book that would an AI be able to convince anybody it was conscious if all mentions of consciousness and feelings and related topics were removed from its training set? And I thought that was a really interesting thing. Would it reach that level of performance without the training?

16:21

Yeah. So we have to remember that all AIs have read the entire human conversation about consciousness and feelings. They've also read novels and poetry, okay? So they're armed with these tools to persuade us they're conscious. So the Turing Test, you know, famously, is like, if a computer could fool an intelligent human behind a screen, that it was another human, then it was intelligent. And computers passed that long ago. But it doesn't work for consciousness because it's very easy to fool us. So the pollen test, which I'm proposing, is an alternative to the Turing Test is build an AI from the bottom up. No mention of consciousness. Can't read any novels, any poetry, any discussion of feeling, and then have a conversation about consciousness. And I'm guessing it wouldn't do very well. No one has yet taken me up on this, but I'm hoping it will happen.

16:39

We do get ourselves into this position where it's tricky either way, because we either may ascribe consciousness in something that isn't conscious, or we may miss it when it is there. Okay. We have to be open to that process either way. I'm not sure how we will know.

17:33

Yeah, I think it's going to be a real issue going forward. And the people who do believe these machines are or will become conscious, they're talking already in terms of the moral consideration we owe these machines, that we are creating an entity that could suffer, and perhaps we have to give it rights. And I think that's really dangerous. I think when we give rights to computers, we'll lose control of them completely. I mean, we gave rights to corporations, right? Corporations have personhood. That hasn't worked out very well. So I think we should be really careful. I'm a little struck by this moral sensitivity of these people in Silicon Valley who are worried about the moral standing and the ability of computers to suffer. I mean, have they stopped eating animals? You know, I mean, like, many humans are receiving minimal moral consideration, and we're already worrying about the machines.

17:52

I want to come back to our own consciousness and ask you whether you think it's actually under threat in the modern world because we have so little opportunity, or we make so little time for our own thoughts these days. I wonder if that is. Is a damaging thing.

18:53

I think it is. I mean, we have so many distractions, and we live in a world, too, that protects us, you know, in the way that animals have to be conscious because they can be eaten. We have technology to protect us. We have the whole superstructure of civilization that allows us to check out. And I've been thinking a lot in terms of the pollution of consciousness, which I think is going on and is widespread, and consciousness hygiene, like, what steps can we take?

19:11

Is your sort of top tip then basically put down your mobile phone or what is it?

19:37

Well, I think that would be the first thing. I mean, think about the way we live now. If you find yourself online, you're at the cafe and you're waiting for the barista to foam your drink. Two minutes of boredom, intervene suddenly, and what do we do? We reach for our phone. What did we used to do, we used to permit ourselves to be bored. But that boredom is like generative. Suddenly you'll start thinking, spontaneous thought will occur and you'll think about what you want to have for dinner or what you're going to be doing that day, or you'll just mind wander. And something creative often comes out of that. So we're choking off that process because we can't stop reaching for our phones. So, yeah, I would say that's number one. The other thing, I think meditation is a really useful way to get in touch with your consciousness and watch your mind at work and just how strange it is. And this isn't just a Buddhist practice. This goes back. I mean, people have been like using introspection to understand consciousness for a long time. David Hume famously, you know, was trying to understand the self and he went into his mind and this is a great exercise to perform, I think, looking for who was thinking his thoughts and. And he found there was nobody home. You have thoughts that are essentially thinking themselves. And I think we just don't stop to consider our mental processes. We proceed. We're very pragmatic, but when you do, you realize the mind is a very strange place to visit.

19:42

Michael Poland, thanks very much for coming on.

21:17

Thank you, Ian. A pleasure talking to you.

21:19

Thanks again to Michael Pollan. To support the Guardian, you can order a World Appears from the Guardian Bookshop. Just head to guardianbookshop.com and before you go, I'd like to recommend the Guardian's brand new investigative series Off Duty. It's about a murder in Chicago and one man's 12 year fight to prove his innocence. To listen to the series, search for Off Duty or Guardian Investigates wherever you listen to your podcasts and hit subscribe. And that's all from us today. This episode was sound designed by Joel Cox and the executive producer was Ellie Burey. We'll be back on Thursday. See you then.

21:23

This is the Guardian.

22:15

Here's how to stay alive longer so you can enjoy Boost Mobile's unlimited plan with a price that never goes up. Do not mistake a wasp nest for a pinata. Stay alive and switch now at boost mobile, after 30 gigs, customers may experience slower speeds. Customers will pay $25 a month as long as they remain active on the Boost Mobile unlimited plan.

22:25

This episode is brought to you by Progressive Insurance. Do you ever find yourself playing the budgeting game? Shifting a little money here, a little there and hoping it all works out well with the name your price tool from Progressive, you can be a better budgeter and potentially lower your insurance bill, too. You tell Progressive what you want to pay for car insurance and they'll help you find options within your budget. Try it today@progressive.com Progressive Casualty Insurance Company and affiliates Price and coverage match limited by state law. Not available in all states.

22:40

Knock knock.

23:09

Ooh, who's there?

23:10

A Boost Mobile expert here to deliver and set up your all new iPhone 17 Pro designed to be the most powerful iPhone ever.

23:11

You call that a knock knock joke?

23:17

This isn't a joke. Boost Mobile really sends experts to deliver and set up your phone at home or work.

23:18

Okay.

23:23

It's just that when people say knock

23:23

knock, there's usually a joke to go with it.

23:25

Like I said, this isn't a joke.

23:26

So the knock knock was just you knocking?

23:28

Yeah, that's how doors work.

23:31

Get the new iPhone 17 Pro delivered and set up by an expert wherever you are. Delivery available for select devices purchased at boostmobile.com, terms apply.

23:32