Latest / Uncomfortable Conversations with Josh Szeps / “Conspiracies, Echo Chambers & Culture Wars” with Sacha Judd

Transcript

“Conspiracies, Echo Chambers & Culture Wars” with Sacha Judd · Uncomfortable Conversations with Josh Szeps

• 0:00
  Get a humans, welcome to the safe space for dangerous ideas.
• 0:07
  This is a cracker of an episode.
• 0:08
  I loved it.
• 0:09
  I get into constructive arguments with a very smart person about free speech and transphobia
• 0:15
  and artificial intelligence and whether Twitter is killing us all.
• 0:20
  What's not to love?
• 0:21
  But first, the Australian election is worth just briefly touching upon.
• 0:25
  If you want to hear a more extended dialogue between myself and another smart, funny Australian
• 0:30
  political analyst, Peter Van Oncel, and on the morning after the election, then there's
• 0:35
  a link in the description to this podcast to the Substack Live, which we did.
• 0:39
  And in general, even if you don't want to pay money for Substack, you should certainly
• 0:42
  subscribe to the Substack in order to be notified when I go live with such illustrious
• 0:48
  people about pressing topical concerns that are probably moving too fast to be able to
• 0:52
  address on this podcast.
• 0:55
  But a lot of you have asked my thoughts about the Australian election and I've received
• 0:58
  some queries from abroad about whether or not this was a repudiation of Donald Trump.
• 1:02
  So let me just say a few words very briefly.
• 1:06
  The brief takeaway from the election, which was a massive landslide towards the government,
• 1:11
  the centre left government, is less than it was towards the government than it was away
• 1:16
  from the centre right opposition, which ran a bumbling and confused campaign that was
• 1:22
  sometimes trumpy and sometimes not trumpy.
• 1:25
  And it was led by a deeply uncharismatic, unlikeable and sort of not particularly physically
• 1:31
  attractive, although that shouldn't matter.
• 1:32
  But of course it does, person named Peter Dutton, who never managed to connect with the Australian
• 1:37
  people or particularly resonate.
• 1:39
  It was a battle of two-dour, middle-aged, white, straight males, the pale, grey totems
• 1:47
  of their parties, neither of them with particularly big ideas.
• 1:50
  But what's interesting about the scale of the electoral victory is, whilst I think it's
• 1:57
  overkill to interpret it as a repudiation of Donald Trump, he didn't really didn't feature
• 2:01
  much in the Australian election campaign at all, it is I think a repudiation of an
• 2:09
  anti-status quo, anti-encumbent energy that has flared up most remarkably in the form
• 2:16
  of Donald Trump.
• 2:18
  And I don't think you would have this election result where it not for Donald Trump.
• 2:22
  So there's a kind of a subtext to the whole thing, which is, and look, Australia, medium-sized
• 2:28
  countries that are a long way from everything have the benefit of watching what everybody
• 2:32
  else does and then choosing to either emulate it or not.
• 2:35
  We see this time again in Australia, where it looks at the school shootings and the nature
• 2:40
  of America's relationship to guns, and when there's a mass shooting here goes, okay,
• 2:45
  well we see what happens when you remain a wash with guns, so why don't we take a different
• 2:49
  path and it does serious gun reform, or on the pandemic as well, where Australia had
• 2:54
  the luxury of copying it just enough later than other major countries that we were able
• 3:01
  to close the borders and mostly live a free and happy virus-free existence until, as we
• 3:08
  know, the calamity of the vaccine delay led to a 100-day plus shutdown in New South Wales.
• 3:13
  And of course, Victoria had all of its own problems, but setting all that aside, the central
• 3:16
  point remains that in Australia, we sort of enjoy gazing with a sense of vacant, distant
• 3:24
  superiority on the rest of the world and then carving our own path on the basis of which
• 3:29
  model we like the best.
• 3:32
  And we've watched Brexit and we've watched the rise of the far-right in Europe and we've
• 3:36
  watched the struggles in Italy and France and we've watched Donald Trump in the United
• 3:40
  States.
• 3:42
  And whilst it hasn't been an explicit part of this election campaign, one can't help but
• 3:47
  conclude that the subtext to this election was, we've seen what happens when you get
• 3:53
  too cute with change.
• 3:56
  When you get too complacent about the good stuff you've already got, when you try to shake
• 4:02
  things up and get super edgy and be all anti-elites and try something brand new because everything
• 4:09
  is so boring these days and what about inflation and bubble-bubber her emails?
• 4:15
  That didn't fly this time round.
• 4:18
  And what we saw, it was not so much a swing to the left because the further left parties
• 4:23
  like the Greens did not overperform.
• 4:25
  What we saw was a rallying around normality, a rallying around the boringness of the existing
• 4:33
  government.
• 4:34
  There is nothing inspiring about this government.
• 4:36
  There is nothing revolutionary about this government.
• 4:38
  There is nothing particularly interesting about this government.
• 4:41
  And Australians were like, yes, in this world, at this moment, that is what we want, something
• 4:47
  not particularly interesting and not particularly threatening better the devil we know.
• 4:53
  That is my takeaway from this election.
• 4:55
  So it wasn't about Donald Trump but it was about the forces that gave rise to Donald Trump
• 4:59
  and Australia saying no to those which heartens me greatly even though I have no particular
• 5:05
  affection necessarily for this government and its small bore thinking.
• 5:10
  We'll see what they do over the coming term.
• 5:12
  Today's guest is Sasha Judd.
• 5:14
  I really, really, really think she's a fabulous and interesting person.
• 5:18
  I think of her actually as a kind of a mark Cuban of New Zealand by which I mean she's
• 5:25
  an entrepreneur and business, you know, high-flying business person and an advisor to government
• 5:33
  and a commentator on big tech and a left-wing progressive.
• 5:38
  And someone who advises startups and venture capital and so on and thinks about the ways
• 5:44
  that companies can grow, particularly in the tech sector and the ways that big tech companies
• 5:48
  have gone wrong.
• 5:50
  She is an expert, ostensibly, on internet subcultures and fandom meaning, you know, the rise
• 5:58
  of fans subcommunities, how they function, how they operate.
• 6:02
  But this also feeds into the question of conspiracy theories and like the meaning that we derive
• 6:07
  from conceiving of ourselves as sub-collectives, whether we are united by our fandom of particular
• 6:13
  celebrities or united by our belief in QAnon or vaccine theories or whatever else it might
• 6:18
  be, COVID conspiracies and so on.
• 6:22
  Sasha says that she wants to bring back the good internet and internet the way that it used
• 6:26
  to be fueled by genuine community and creativity and connection, whether or not that internet
• 6:33
  ever really existed is the subject of this conversation and the kind of internet that
• 6:37
  we might build in the future is very much as well.
• 6:40
  I hope you enjoy as much as I did, the one, the only, Sasha Judd.
• 6:46
  It's called the good internet and it's something I've been thinking about this year about how
• 6:53
  we can try and reclaim some of our digital agency and get away from this sort of handful
• 6:59
  of wall gardens that we find ourselves living in online.
• 7:02
  Yeah, what are the handful of wall gardens that we're living in online?
• 7:06
  Well, I think we've been forced into these tiny, crowded, noisy platforms that are owned
• 7:11
  by the likes of Metta, Elon Musk and others and I think it's devolved into this situation
• 7:20
  where we've lost our digital agency, the web is an infinite place that we used to build
• 7:27
  our own spaces on and over time, those have been consolidated down and we've moved more
• 7:32
  and more to being on the same handful of platforms as everybody else.
• 7:36
  Did we ever build our own spaces on it or is that a healthy on nostalgia in the sense
• 7:41
  that when we always on, we started out on my space and then we migrated to Facebook,
• 7:44
  we would never, I was never coded.
• 7:46
  I was never building websites, you know.
• 7:50
  Right.
• 7:51
  Well, that's an interesting thing to think about.
• 7:54
  I started out on Usenet, which was a pre-web proto-reddit, I guess, a sort of bastard love
• 8:00
  child between email and a bulletin board, but anyone could create a Usenet group around
• 8:06
  something that they were interested in.
• 8:08
  And so the first places we went online, I think regardless of when you came online, the
• 8:12
  first thing you did was look for people who had shared interest, who were into the same
• 8:17
  things that you were, whether that was a sports team looking for results or a movie
• 8:21
  that you'd just seen that you were interested in or whatever.
• 8:24
  So you went looking for a neighborhood with people who were interested in the same things.
• 8:28
  I was mostly looking for friends who I already knew in the physical world and pornography.
• 8:33
  Those were the two.
• 8:34
  And I think that's not uncommon.
• 8:35
  Well, the pornography is certainly not uncommon.
• 8:37
  It's routine age-lad.
• 8:38
  I do think that now there's much more focus on finding people you know online.
• 8:43
  I think this generation is coming online and the people they know on social are people
• 8:47
  they know in real life.
• 8:48
  Yeah, that doesn't even know each other's phone numbers anymore because they just text
• 8:51
  each other on Instagram, DMs.
• 8:52
  It's like, you know, what's Sarah's number, I don't know, how do you communicate on
• 8:57
  meta?
• 8:58
  Everything's on meta.
• 8:59
  But back in the day, we made our own spaces.
• 9:01
  And so one of the things that I am thinking about and talking about and writing about a lot
• 9:05
  is online communities and particularly fan communities because fan ups had a really
• 9:10
  long history of making their own spaces.
• 9:14
  Fandom.
• 9:15
  Fandom.
• 9:16
  What do we mean by fandom?
• 9:17
  Yeah, good question.
• 9:18
  So there's different kinds of fandom.
• 9:21
  When I talk about it, I'm talking about what we call transformative fandom, which is communities
• 9:25
  of people who don't just consume media, but debate it, reimagine it, write stories, create
• 9:31
  art about it, expand its universe.
• 9:34
  Transformative fans are often compared and contrasted to what's called curatorial fandom.
• 9:39
  And those are sort of your more classic, um, fans who are interested in assembling the
• 9:45
  most facts.
• 9:46
  And the people who can tell you every Wallabies captain for the last 10 years for every aspect
• 9:52
  of the Star Wars universe and how it intersects.
• 9:54
  Yeah.
• 9:55
  So those bedrooms are covered in posters of one direction.
• 9:58
  Well, one direction fans tend to be transformative fans.
• 10:01
  Oh, that's interesting.
• 10:02
  What makes them that?
• 10:04
  Because they're engaged in creation.
• 10:06
  So stories, art, photo edits, video cuts, fan edits, that's a huge part of Godoucher.
• 10:14
  Including conspiracy theories.
• 10:15
  I just pulled one direction out of my bum, but you do have a whole, I mean, I've heard
• 10:19
  you talk about how the conspiracy theory about the two singers in one direction, actually
• 10:25
  having an elicit gay romance is an interesting model of the way that fan communities and
• 10:30
  especially creative ones can come to obsess about minor details and drive meaning from their
• 10:36
  in-group status and in a particular conspiracy.
• 10:39
  Can you just tell people who don't know about that?
• 10:41
  What that is?
• 10:42
  Yeah, sure.
• 10:43
  You know, I used to think that celebrity conspiracy thinking was kind of cute and funny.
• 10:47
  And I think a lot of us are sort of drawn to the idea of, you know, was Beyonce ever
• 10:51
  pregnant or is Taylor Swift secretly gay and, you know, hiding lyrics in her songs.
• 10:58
  And there's a lot of that in pop culture fandom, this sort of search for signs and signals.
• 11:04
  And I've just actually published a long piece on this, this week on the fan's planning
• 11:09
  website called the RPF question, an RPF stands for real person fiction.
• 11:13
  So it's fandom around celebrities, artists, actors and so on, rather than the characters
• 11:20
  that they play.
• 11:21
  I see.
• 11:22
  And so people who write stories, imagine relationships between people who are real people.
• 11:28
  And that started in the early 2000s with band fandom.
• 11:33
  So particularly around pop web bands like Insink and the Backstreet Boys and emo.
• 11:39
  Just to clarify, the conspiracy theory is about if you play a rock song backwards, there
• 11:44
  will be a satanic that, you know, it'll make it have a satanic message or that beetles
• 11:48
  are actually involved in this or that.
• 11:50
  Paul's really dead or whatever.
• 11:52
  Yeah, or, you know, exhale, or that Elvis, you know, didn't die or that JFK was having
• 11:57
  an affair with Marilyn Monroe, some of these true, some of these untrue.
• 12:01
  Why do those not fit into the same basket of thinking as what, as what you just articulated?
• 12:06
  You said it started in the 2000s.
• 12:07
  Oh, I was talking specifically about real person fiction.
• 12:12
  So that's a, that's a fandom concept, which, okay, got it.
• 12:15
  Okay.
• 12:16
  Not a long history.
• 12:17
  Right.
• 12:18
  Right.
• 12:19
  Right back to, you know, the tabloid magazines of the 20s and 30s that were sort of
• 12:22
  encouraging narratives about studio stylists and whether they were, you know, probably
• 12:28
  back to the pharaohs and whether Cleopatra was, like, in this person or who knows?
• 12:33
  Nero fiddling while Roan burned.
• 12:34
  Right.
• 12:35
  Just one of our earliest conspiracy theories, so yeah, it's got a long, a long history.
• 12:39
  Okay.
• 12:40
  Sorry, carry on.
• 12:41
  Yeah.
• 12:42
  So I, but I think what's, what's been happening is with all conspiracy thinking, there's
• 12:46
  always a kernel of truth, right?
• 12:47
  We know that people in power lie, we know that some paparazzi shots are staged, like we,
• 12:53
  we know that these things are true and we have to question what we see.
• 12:57
  But unfortunately, I think what's happening is the celebrity conspiracy spills over into
• 13:04
  mainline conspiracy thinking and then it's kind of question everything.
• 13:08
  Nothing is true anymore.
• 13:10
  And I think that takes us to quite a dangerous place.
• 13:13
  And I think more and more people are looking for community online and they're finding it
• 13:19
  in communities that celebrate conspiracy thinking and I think that's dangerous.
• 13:24
  It's interesting that you say that paparazzi shots are staged as one of the examples of
• 13:28
  conspiracy thinking because one thing that I've learned as I've grown older and also more
• 13:33
  connected to powerful people is that there are some areas where I have justification for
• 13:39
  being extremely skeptical and more skeptical than the average person.
• 13:44
  And others where I know that the average person is too skeptical actually.
• 13:48
  Yeah.
• 13:49
  So there are some fields like paparazzi and gossip magazines where I think most people
• 13:54
  even though they sort of know that they're mostly bullshit, don't know how bullshit
• 13:58
  they are and not aware of just how much of a close relationship there is between a movie
• 14:03
  stars agents, the publicists, the publishers of magazines and of online sites, that there's
• 14:09
  this symbiosis going on where nothing is accident, very, very few things are accidental.
• 14:13
  A lot of all of it is pretty much stage managed.
• 14:17
  So I'm very skeptical of that.
• 14:18
  And then when it comes to political and corporate power, but political power is one that I know
• 14:24
  more because I've had more to do with people who wheeled it.
• 14:28
  I think most people have a vastly overinflated estimation of the capacity of politics to pull
• 14:33
  off conspiracies.
• 14:34
  And I regard it much more as a bumbling plan on how to have to be involved.
• 14:38
  Yeah.
• 14:39
  That's right.
• 14:40
  And just not to sort of name drop, but when you've I lived in the States, when you've
• 14:46
  been to cocktail parties with Colin Powell, it just does make you realize that he's just
• 14:53
  a person who goes to the toilet and takes a shit, you know, when he needs to.
• 14:57
  He's not a puppet master who was in like, it becomes increasingly implausible to believe
• 15:03
  the extremes of conspiracy theories in the field of politics and increasingly more plausible
• 15:09
  to believe them in the field of entertainment.
• 15:12
  How is an ordinary, how is a normie supposed to tell the difference between the two?
• 15:16
  Oh, I know.
• 15:17
  And I mean, this is one of the things that I find so alarming about how we prepare the
• 15:21
  next generation.
• 15:22
  Like how do we teach our kids defense against the dark arts because the, the, you know, what
• 15:29
  historically what we've said is that you need to engage in online literacy.
• 15:33
  You should question sources.
• 15:36
  You should make sure that you're doing your own research.
• 15:39
  Well, those are all the mantras of complete cookets, right?
• 15:42
  Like, oh, do your own research and what they mean by that is ignore science and experts.
• 15:48
  So I do think it's a really challenging problem.
• 15:52
  And we, we have started to see some great education efforts out of places like Finland.
• 15:57
  It's always the scandals.
• 15:58
  It's always further down.
• 15:59
  And Finland in particular is like Finland and Denmark.
• 16:01
  Yeah.
• 16:02
  Norway.
• 16:03
  They're doing some incredible programs with improving digital literacy for tens of
• 16:08
  years.
• 16:09
  What do they look like?
• 16:10
  A lot of it is starting early.
• 16:12
  It's baking it into all subjects in the curriculum, rather than just making it a one-off module
• 16:16
  that students are doing.
• 16:18
  And just equipping kids to have a much clearer idea of what they're being fed, what the motivations
• 16:25
  are behind it, and how they can think sort of critically and neutrally about the information
• 16:29
  they're receiving.
• 16:30
  So is it the case that we shouldn't do our own research?
• 16:33
  Haha.
• 16:34
  I absolutely think you should do your own research.
• 16:36
  I think understanding what that means, equality of sources, understanding, yeah, motivation.
• 16:43
  What's the motivation behind the person who's put together the article or the collection
• 16:48
  of sources?
• 16:49
  I mean, it can be very hard to know.
• 16:51
  And also, the motivation can sometimes be innocent in the case of, you know, take vaccine
• 16:56
  misinformation.
• 16:57
  I think most of the people peddling at our innocent are trying to do the right thing.
• 17:01
  Yeah.
• 17:02
  I'm concerned about their children.
• 17:03
  Absolutely believe that.
• 17:04
  Because I think that, you know, if vaccines really did cause deaths at the rate that these
• 17:09
  people believe, we would all have an obligation to stand up for it, but it's just not true.
• 17:18
  And so trying to untangle that sense, and I gave a talk in Germany a couple of years
• 17:24
  ago called everything breaks a scale, and I really believe that.
• 17:29
  I believe that these things start out well-intentioned, that these people are looking for a sense
• 17:34
  of community.
• 17:35
  They're looking for a sense of purpose, and there's kind of a dopamine hit that comes from
• 17:40
  your content being liked and shared, and gradually you're the star of your little Facebook
• 17:44
  group.
• 17:45
  And, you know, in order to continue to receive that validation, you have to sort of have
• 17:50
  more and more content and more and more outlandish theories.
• 17:55
  And, and gradually this just sort of magnifies into a situation where at scale, we're seeing
• 18:01
  this really widespread disinformation being propagated.
• 18:04
  Do you need the at scale in that sort of framework?
• 18:09
  Like isn't it possible to just be a normal everyday person who has an instinct that in general
• 18:17
  power structures are corrupt and misleading and have their own interests at heart that
• 18:21
  drug companies will be in cooots with public health officials and probably don't care
• 18:26
  about your child as much as you do, and therefore you should err on the side of assuming that
• 18:31
  official information is suspect, and do your own research, and then your own research
• 18:35
  leads you into, you know, having an open mind and questioning and so on, and then it becomes
• 18:39
  very difficult to...
• 18:40
  It's a second question.
• 18:41
  Well, yeah, you know, and then it does become difficult to ascertain whether or not the
• 18:46
  barrage of vaccinations that we give young children is correlated to autism or not.
• 18:50
  Like, you know, I trust the experts who tell me that it's not, but that is an act of
• 18:55
  faith to some extent on my part, because I'm temperamentally, well, the amount of research
• 19:01
  that I would have to do, sure, but these people are doing a lot of research, they're
• 19:05
  ignoring the experts, and you know, again, there's always a kernel of truth at the heart
• 19:11
  of it.
• 19:12
  Like, you know, I understand how people wind up on a wellness to conspiracy pipeline.
• 19:18
  I think that the mainstream medicine has in the past made mistakes, it certainly hasn't
• 19:26
  listened to women and people of colour about their health issues in a way that it should
• 19:31
  have.
• 19:32
  People have been dismissed by mainstream medicine, so of course they go elsewhere to
• 19:37
  try and find answers.
• 19:39
  I understand that at the heart of that, there is a kernel of truth, but unfortunately,
• 19:44
  and this is where I come back to the scale part.
• 19:47
  It's one thing, if you're just you, little nor me, having your, you know, concerns and
• 19:53
  asking your questions in your community, but as soon as you thrust that into an online
• 19:58
  city of millions of people and amplify it in increasingly accelerated ways, and the platforms
• 20:07
  know that what drives engagement is conflict and rage and fear, so that's all what they
• 20:13
  are amplifying, then that's when scale takes it outside of your little neighbourhood.
• 20:20
  I'm still, I still struggle with the research thing somewhat.
• 20:25
  I had a public radio show here for, for a time, a talk back radio show and it was during
• 20:31
  the tail end of COVID, and I had a segment called Do Your Own Research where we would take
• 20:35
  a factoid, an untrue fact that was circulating online or a suspect fact, usually about medicine,
• 20:42
  and instead of untelling and say to people, you'll see a lot of people on Facebook saying
• 20:46
  you should do your own research about this.
• 20:48
  So to save you the hassle of doing the research, we've just got the world's best expert
• 20:52
  in this to tell us what the answer is.
• 20:54
  So you don't have to do the research.
• 20:56
  And it was a way of turning that principle on its head to be like, don't do the research
• 20:59
  because what people mean by doing the research is just ingest a huge amount of misinformation
• 21:03
  about this thing.
• 21:05
  But it also meant that we addressed topics that I felt the mainstream media was reluctant
• 21:10
  to address because they were associated with this kind of nasty fringe element, right?
• 21:16
  So there were things like there was an uptick in mortality after COVID and vaccine skeptics
• 21:24
  thought that that was attributable to the vaccines, and the mainstream media really just
• 21:28
  sort of didn't know what to do with it because we didn't really know at the time why more
• 21:32
  people were dying.
• 21:33
  The classic one at the start of COVID was the blood costs associated with one of the
• 21:37
  vaccines.
• 21:38
  And that was a story that I think even slowed down the roll out here in Australia right
• 21:43
  of that.
• 21:44
  Right.
• 21:45
  Yeah, to later.
• 21:46
  Yeah.
• 21:47
  The Moderna vaccine.
• 21:48
  And when you just looked at the bare facts, the rate of blood clots was less than the
• 21:52
  rate that women experience on birth control.
• 21:55
  And we've lived with that risk for 60 plus years.
• 22:01
  But blood clots associated with COVID vaccine is a much six year story, right?
• 22:06
  So, yeah, I agree with you.
• 22:08
  There's a challenge because I think we also have plenty of evidence that platforming misinformation
• 22:15
  has dangerous effects as well.
• 22:17
  And so I think the fear is...
• 22:18
  Even if you're debunking it.
• 22:19
  Oh, it can be.
• 22:20
  It's tricky.
• 22:21
  I mean, I always felt this because I know that there's this effect, right, where people,
• 22:25
  if people hear a claim, even if it's a claim being debunked, it does have a stickiness
• 22:30
  to it.
• 22:31
  And then lodge in their heads and they can sometimes forget whether or not it was true
• 22:34
  or not true.
• 22:35
  So even the raising of it, and I think many journalists have a much more cautious attitude
• 22:40
  about talking about things than I instinctively do.
• 22:43
  So I would always err on the side of, I'll just talk about it.
• 22:46
  I mean, you know, the reason why there was an increase in mortality was multifactorial,
• 22:51
  but it didn't have anything to do with the vaccines because most of those increases
• 22:54
  were from things like dementia and cancer and heart disease, where the causes predate
• 22:59
  the pandemic.
• 23:00
  It had to do with access to hospitals and all kinds of other factors that were related
• 23:04
  to the pandemic, but not to the vaccine.
• 23:06
  So I felt like just hearing an expert say that might lead some people to think, oh, there's
• 23:13
  this controversy about the vaccine being deadly, but I felt that that was outweighed by
• 23:18
  the number of people who would go, oh, I saw that on the Facebook, it's good to know
• 23:22
  that was bullshit.
• 23:23
  Yeah.
• 23:24
  Because you're right.
• 23:25
  The misinformation spreads, I mean, I think one of the classic examples we've seen
• 23:29
  over the last couple of years is that insane story about Irish.
• 23:32
  There's always some school that's put letterboxes in for their children who are, what do you
• 23:36
  start showing?
• 23:37
  What the fuck with that story?
• 23:38
  I hear it so many people say it, and I can't get back to the source of when that ever happened
• 23:42
  to me.
• 23:43
  Well, there's an NBC disinformation journalist who tracked it all the way back.
• 23:46
  Just tell people the story in case they're not as online as well.
• 23:49
  So this is one of those, in the old days, we would have called it an urban legend, right?
• 23:53
  It's the sort of thing that you would have looked up on snopes.com to say if it was incredible
• 23:57
  or not.
• 23:58
  I have had three credible people claim that it's their friends.
• 23:59
  Oh, they're friends.
• 24:00
  Kids, kids, their friends, mothers, aunts, whatever.
• 24:04
  And the story goes that at this person's school, the school has put in place letterboxes
• 24:11
  for kids who identify as cats.
• 24:15
  And I mean, it's a classic sort of transpanic urban legend.
• 24:19
  Yep.
• 24:20
  So the idea is that schools have gone so woke that not only are they encouraging or allowing
• 24:24
  children to switch genders, they're allowing them to identify as whatever they want to,
• 24:29
  including felines.
• 24:30
  And they're accommodating.
• 24:31
  And it feeds into all sorts of, it feeds into bathroom panic, it's just the worst example
• 24:36
  of that kind of bullshit urban legend.
• 24:41
  And it's never happened anywhere, and yet it spreads, and it gets picked up by the mainstream
• 24:48
  media every time it sort of migrates to a new place.
• 24:51
  And so it had been in the US for a while, as I say, this NBC journalist had gone right
• 24:55
  back to debunk it, no one's ever been able to prove that it ever happened.
• 24:59
  And then six, eight months later, it's popped up in England, it's a school in Scotland,
• 25:03
  it's a school in Wales.
• 25:05
  And then next minute, it was in New Zealand, it was a school in Funga Rae up near where
• 25:09
  I live.
• 25:10
  And I have credible, professional contemporaries saying to me, oh, did you hear about that
• 25:17
  school?
• 25:18
  Me too.
• 25:19
  Me too.
• 25:20
  And they're not claiming that they've heard a rumour.
• 25:22
  They're claiming that it happened to a friend of their kids.
• 25:26
  We've lost, we've lost out, but yeah, I mean, I'm less polyannar about what the past
• 25:31
  used to be like.
• 25:32
  I think people will probably completely riven with all kinds of nonsense.
• 25:35
  Oh, look, I'm not polyannar about it at all.
• 25:37
  I, you know, the early internet was a very small, very homogenous space.
• 25:41
  But even before the, I mean, in the 1950s, people probably had, well, all kinds of stupid
• 25:45
  nonsense.
• 25:46
  And the satanic panic of the 1980s, Kennedy assassination.
• 25:49
  Yeah.
• 25:50
  It's snatched off the streets by crazed satanic perverts, like, you know, the people are susceptible
• 25:56
  to this.
• 25:57
  But I think what has happened, what the internet has done is, is provided a way to amplify
• 26:03
  these things at a speed and a scale that we haven't seen before.
• 26:07
  So, you know, I was super into the Kennedy assassination when I was a teenager.
• 26:11
  I thought it was so interesting.
• 26:12
  But if I wanted to read Crackpot conspiracy theories about who called Kennedy, I had to
• 26:16
  go and find out a print books in the library or the bookstore, you know, like, it wasn't
• 26:21
  being forced fed at me by packaged TikToks that were giving me, you know, a little sound
• 26:28
  bite that I could tell you.
• 26:29
  What do you believe about it now?
• 26:31
  I mean, to your point earlier, like, the idea of the number of people who would have to
• 26:35
  have been involved in kept at secret all this time, it's just, you know, or not, maybe
• 26:40
  not.
• 26:41
  Like, this is one that I'm on the fence on.
• 26:42
  I really don't know.
• 26:43
  I mean, I just believe whatever the last person who spoke to me about it, who knows
• 26:46
  more about it than I did.
• 26:47
  Oh, sure.
• 26:48
  And there's some amazing stories.
• 26:49
  It's really weird.
• 26:50
  It's fucking weird, Sasha, that the person, that the assassin gets assassinated a few days
• 26:56
  after, and pulling off this lone assassin thing.
• 26:59
  It's like, it's strange.
• 27:02
  And it's not inconceivable.
• 27:04
  But there's something lovely about it.
• 27:05
  That there's a cabal of only three people in jail, right?
• 27:08
  They tickle something in our brain.
• 27:10
  Totally.
• 27:11
  Yeah.
• 27:12
  It's fascinating.
• 27:13
  It's like you always want that sort of mystery to solve.
• 27:16
  Well, also, when you are characterizing the archetype of the person who, you know,
• 27:21
  gets their rocks off by building a following of misinformation or conspiracy thinking online,
• 27:27
  there is a more sort of generous interpretation of that psychological mindset, which is
• 27:33
  that their contrarians by nature, I'm sympathetic to this because I am a bit as well.
• 27:40
  And they just don't quite know where to stop.
• 27:44
  Like I do think it's useful to play devil's advocate.
• 27:48
  Well, yeah, and maybe more than play devil's advocate always, because play devil's advocate
• 27:52
  assumes that you're doing it insincerely or for a strategic purpose.
• 27:56
  But to genuinely think there's a utility in questioning quite strongly, receive wisdom
• 28:05
  about things.
• 28:07
  Because obviously we all know that in every place and every time the majority of people
• 28:13
  have a lot of blind spots, we can see them when we look at other places.
• 28:17
  We can see them when we look into the past.
• 28:19
  It's a certainty that in the future and in other places, people will look at our society
• 28:22
  and our forms of governance and see things that we can't see right now.
• 28:27
  Yeah.
• 28:28
  So it does make a certain sense temperamentally for a particular type of person to be like,
• 28:32
  no, I mean, I'm going to scratch at these wounds.
• 28:36
  And I'm going to figure out what's underneath.
• 28:38
  Yeah.
• 28:39
  And I think utopians at the start of the digital age were very much of that view.
• 28:43
  That what the internet brought us was access to all the world's information and that sunlight
• 28:48
  was the best disinfectant.
• 28:50
  And you know, I have a huge amount of sympathy for that view.
• 28:52
  I was a university debater on 15 minutes notice.
• 28:56
  I could argue any side of any topic.
• 28:59
  But I think what the marketplace of ideas has taught us over the last 20 years is that
• 29:04
  there's a market for truly terrible ideas.
• 29:07
  And that's why measles is about to become endemic in the United States again.
• 29:11
  Yeah.
• 29:12
  I mean, I think the elephant in the room here is the algorithm, right?
• 29:16
  Is the fact that this landscape, this terrain of information and conversation is not a neutral
• 29:24
  space.
• 29:25
  No, it's not a.
• 29:26
  There's no such thing as a neutral space.
• 29:28
  Well, even if it, even if there were, I mean, yes, of course.
• 29:32
  But I mean, there are better and there are more and less neutral spaces and to have software
• 29:38
  engineers in Menlo Park, figuring out what's going to be the stickiest, clickiest, shareiest,
• 29:45
  likeiest meme.
• 29:48
  And that that's the only measure of whether or not a piece of information or content is
• 29:53
  elevated or downplayed, particularly when we now know that there's a bonkers way to run
• 29:59
  a civilization.
• 30:01
  And we now know that the things that are driving the most engagement are negative.
• 30:05
  Like this isn't a balanced thing though.
• 30:08
  It's hatred, conflict, fear.
• 30:11
  That's what drives engagement.
• 30:12
  Are they negative in the way that you?
• 30:15
  Are they negative in their valence or are they just negative in their consequence?
• 30:19
  By which I mean, I often hear it claimed that social media is a toxic place and that
• 30:26
  you never hear ideas that you disagree with on social media because it reinforces your
• 30:31
  own opinions.
• 30:33
  But those seem to be slightly contradictory to me.
• 30:36
  Like are you only seeing things that you agree with or is it a toxic place where you're
• 30:39
  getting yelled at?
• 30:41
  It strikes me that what's actually happening is that you're seeing because anyone can
• 30:47
  open up their social media feeds and they can see a bunch of things that they disagree
• 30:50
  with.
• 30:51
  Sure.
• 30:53
  It's just that the versions of the things that they disagree with will be primed to outrage
• 30:59
  them rather than being the most reasonable and nuanced version of things that they disagree
• 31:03
  with.
• 31:04
  Sure.
• 31:05
  Because they're sound bites.
• 31:06
  Yeah.
• 31:07
  Well, because that's likely to make them linger on it or comment on it or share it, right?
• 31:11
  Yeah.
• 31:12
  To engage with it.
• 31:13
  Yeah.
• 31:14
  And the things that reinforce what they believe will be super placating to their existing
• 31:21
  world view rather than nuanced and challenging it and, you know, on the other hand, D. So is
• 31:30
  the experience of being on social media negative or is it just that social media is negative
• 31:35
  even when it feels positive, if that makes sense?
• 31:40
  I mean, I do, to come back to your earlier thing, I do think it is about we've lost the
• 31:44
  ability to control our scroll, right?
• 31:46
  We, you know, an early sort of cliche for online life used to be that you needed to cure
• 31:55
  at your own media experience.
• 31:56
  You know, like you should just follow people that you are going to get great information
• 32:01
  from.
• 32:02
  It's up to you to block or to, you know, not feed the trolls, etc., etc.
• 32:07
  But we actually have lost the ability to do that because these platforms that we're
• 32:11
  on just want to throw for you or are recommended or are you might like at you constantly, so
• 32:19
  you don't have the ability, you can't cure at TikTok, like it's just going to throw things
• 32:23
  at you.
• 32:24
  Yeah.
• 32:25
  And so without that control, then I think it does become an extremely negative place because
• 32:31
  you don't have any ability to do that.
• 32:35
  And I think that's one of the super interesting things that's changing right now.
• 32:38
  So Blue Sky, which is Twitter replacement for one of a better word, they've been growing
• 32:45
  exponentially since Elon Musk acquired Twitter.
• 32:48
  And when I joined up when he acquired it, and at first when I looked at Blue Sky, I thought
• 32:55
  it's nostalgic, but it's unlikely to be the future.
• 32:58
  Like we didn't know we needed Twitter when we built it, whatever comes next is not going
• 33:02
  to be Twitter clone.
• 33:04
  But now I'm not so sure, I think that one of the things that's super interesting about Blue
• 33:08
  Sky is that it lets you create your own custom feeds.
• 33:13
  And unlike old Twitter, which let you make lists, so you could check in on a group of accounts.
• 33:18
  A custom feed can be, you can mute or block words or accounts, you can make sure you see
• 33:25
  every post on a particular topic or by a particular person.
• 33:29
  You can weed out Amazon links, you can moderate, not say for work content, you can make sure
• 33:35
  you see every video of Harry Styles that appears.
• 33:38
  You can create a feed that is just exactly what you want your social media experience today.
• 33:43
  How do I create the Harry Styles feed?
• 33:45
  Just asking for a friend.
• 33:46
  Look, I've already made it, so you can just follow mine.
• 33:49
  Fabulous.
• 33:50
  And so I originally in the early days of Twitter looked with scorn on people who blocked
• 33:56
  a lot.
• 33:57
  I felt like the whole point of social media should be to expose me to a common experience
• 34:02
  that like we should as much as possible be trying to strive towards a unified experience
• 34:06
  on social media.
• 34:08
  And there was something electrifying about early Twitter, about 2012 to 2014 Twitter, I
• 34:14
  suppose.
• 34:15
  It was much smaller for status.
• 34:16
  Yeah, but it also felt, and maybe it was just because I was living in New York and
• 34:20
  I was a journalist, but it felt like everyone in Maddard was on it.
• 34:22
  Everyone who was influential in journalism and politics was there talking about, and
• 34:28
  it was a place where news would break, it was a place where you could go to get and verified
• 34:32
  people who you trusted would be delivering information that was credible.
• 34:37
  And people didn't immediately pile onto that and get into arguments in the comments quite
• 34:40
  so much.
• 34:41
  And I had friends who would just block, block, block so that they, because they are not
• 34:45
  putting up with this shit.
• 34:46
  And I thought that really contravenes the spirit of this enterprise.
• 34:51
  Unless someone was overt, I mean, I might have blocked five people in my entire life.
• 34:55
  They would have to be overtly threatening and, you know, racist or sexist or homophobic
• 35:01
  or something.
• 35:02
  You know, them in the most vile possible way.
• 35:04
  Do you think your experience was a little bit different as a white man?
• 35:08
  Entirely possible.
• 35:09
  Entirely possible.
• 35:10
  I am a gay man, and I'm a Jewish man, so there is, I'm not immune entirely, but it's
• 35:16
  been absolutely certain that if I was a woman of color, that would be a different experience.
• 35:21
  But I still felt that a lot of the blocking was not done because people were receiving
• 35:27
  sexist or racist slurs.
• 35:28
  It was being done because people wanted to curate their own walled garden of like-minded
• 35:33
  people.
• 35:34
  It was people, you know, a lot of my colleagues were blocking people.
• 35:38
  What do you think I would gain from seeing Alex Jones in my feed?
• 35:43
  Well now nothing.
• 35:44
  I mean, I'm not making the case that your policy in 2025 should be the same as my policy
• 35:48
  in 2012.
• 35:49
  I'm sort of saying-
• 35:50
  A lot has evolved since then.
• 35:51
  Yeah.
• 35:52
  I'm saying what has happened?
• 35:54
  Was that a naive moment?
• 35:58
  Because a lot of them blocking it struck me with just to clarify that and to close the
• 36:01
  loop on it was, I don't have to listen to assholes who are talking about why, you know,
• 36:10
  I don't know, Muslims need to be integrated better into society or so.
• 36:14
  So it'll be-
• 36:15
  It'll be something that would be adjacent to a toxic opinion, but was not actually in
• 36:20
  itself a toxic opinion, right?
• 36:22
  It is fundamentally a political disagreement or, you know, the rate of immigration, for
• 36:25
  example, or something like that.
• 36:27
  But an issue is that people didn't want to be bothered actually engaging with.
• 36:30
  They were able to insulate themselves from.
• 36:34
  And my worry is that in a world of echo chambers where everybody is insulated from ideas
• 36:39
  that they regard as being beyond the pale, you just stand up with an unmanageable
• 36:44
  demos of people who don't understand how the people around them think.
• 36:47
  That was my concern at the time, it remains my concern.
• 36:51
  How do you think about that?
• 36:52
  Yeah.
• 36:53
  It's a great challenge because my theory is shoving everyone together in these single
• 37:00
  noisy crowded spaces is part of the problem, right?
• 37:03
  Like cities are inhospitable places and the way that we make them livable is by finding
• 37:10
  vibrant neighborhoods, by finding clubs and social groups and so on.
• 37:14
  And I think that my dream for the internet is that we get back to creating some of those
• 37:19
  neighborhood spaces ourselves.
• 37:21
  And then almost immediately you go, well, isn't that just an echo chamber?
• 37:24
  Like, I don't you just want to hang out with people who look and sound and so on like
• 37:28
  you.
• 37:29
  And I think it's a great- it's a great challenge.
• 37:31
  I'm not sure what the answer is.
• 37:33
  I do think that there's nothing to be gained for me in subjecting myself to people who are
• 37:41
  trying to be edge lords or who are repeating or amplifying ideas that I find unpalatable.
• 37:51
  I don't think that means that I'm abdicating from learning things.
• 37:55
  One of the things I loved about old Twitter was the ability to go and find an expert on
• 38:00
  any topic and follow them.
• 38:02
  I mean, there's been nothing like that before or since also not just follow them actually
• 38:07
  talk to them.
• 38:08
  I mean, what are the great losses for me of Twitter going down, you know, the sewer is
• 38:13
  I could actually communicate with people because I could DM them.
• 38:16
  And if you had a certain amount of credibility on Twitter and a certain number of followers
• 38:20
  and maybe a verified blue check mark, that you could I could ask guests.
• 38:24
  It's so much harder to book guests for this show now because I don't know how to reach
• 38:28
  out to people.
• 38:29
  You know, my friendship with Joe Rogan began because I just DMed him and, you know, random
• 38:34
  people who are stratospheric.
• 38:37
  You could strike up relations.
• 38:38
  Yeah.
• 38:39
  And any time like I remember when the Canadian residential schools stories were breaking
• 38:42
  in the Canadian media and I thought, you know, that's something I don't know anything
• 38:46
  about.
• 38:47
  And within minutes, I was able to find half a dozen indigenous forces on Twitter to follow
• 38:52
  to get background and history and start to understand, you know, the importance of that
• 38:56
  story and, like, for free and, you know, it was, it was an amazing resource that's been
• 39:03
  lost.
• 39:04
  I know.
• 39:05
  So, you know, I do believe there's something we need to save in that.
• 39:10
  But we have to do it in a way that doesn't turn these spaces into, you know, I'm sure
• 39:14
  you're familiar with the Nazi bar story.
• 39:16
  It was like, we can't turn them into Nazi bars.
• 39:19
  Yeah.
• 39:20
  I'll come to that, but I just want to drop the metaphor of a neighborhood.
• 39:23
  Sure.
• 39:25
  One thing that I love about a neighborhood is it's unexpected in us and it's eclectic.
• 39:31
  Serendipity.
• 39:32
  Serendipity.
• 39:33
  Yeah.
• 39:34
  I mean, that's one of the great things about the real world instead of the online world,
• 39:36
  right?
• 39:37
  Yes.
• 39:38
  It's less, I do think it's less predictable.
• 39:39
  Like when I lived in Brooklyn, yes, it's a very, look, it's both homogenous and also heterogeneous.
• 39:52
  So it's homogenous in the sense that you have to have a bit of money to be able to live
• 39:56
  in Brooklyn.
• 39:57
  And you have to want to live in Brooklyn.
• 40:00
  It's ethnically and racially very diverse.
• 40:03
  But I think we can make too much of that sometimes because all of the kinds of ethnically
• 40:08
  and racially diverse people have the same political point of view, the same attitude towards
• 40:13
  racial justice, the same attitude towards, you know, politics, I mean, there is a group
• 40:18
  think there as well.
• 40:20
  But I loved the ability to sit in the bar and not really know who was going to walk in
• 40:25
  the door.
• 40:26
  I mean, yes, it wasn't going to be probably a Trump voting trucker, right?
• 40:32
  But it could be someone who disagreed with it.
• 40:34
  Absolutely.
• 40:35
  It certainly could have been an Uzbekie, you know, Uber driver or anything.
• 40:41
  And my worry is that we've given up on replicating that online.
• 40:46
  Yeah.
• 40:47
  I worry about that too.
• 40:48
  And so, you know, I think one of the things that I've been wrestling with this year is
• 40:51
  I think about how we get the good internet back is how we stop thinking about audience
• 40:56
  and reach, which are sort of functions of social media at its worst and start thinking
• 41:03
  about discovering, like, searches now been destroyed by our generated content.
• 41:08
  So how do we find the vibrant neighborhoods, like how do we find the healthy online communities,
• 41:15
  the places that people still really enjoy going to?
• 41:18
  And I think one of the responses to the toxicity of social media is that people have drawn
• 41:22
  back into smaller, often private spaces.
• 41:26
  So, you know, if you ask someone where their favorite place on the internet now is it's
• 41:30
  likely to be a group chat, a discord about a game or a podcast, a comment section on
• 41:40
  the subject or somewhere that they have found a community of people and it might have started
• 41:45
  out about something specific, but they've built their own little neighborhood and I'm
• 41:53
  really interested.
• 41:54
  But is it a neighborhood or is it an echo chamber, because the neighborhood, the thing that
• 42:02
  I like about the Brooklyn bar is that it's not that I didn't create it.
• 42:08
  I sort it out, but I'm not policing it in any way and it is intentionally eclectic,
• 42:13
  like sure.
• 42:14
  But don't you think that the fans of your podcast, for example, if you started a discord
• 42:19
  tomorrow and said, right, everybody who is interested in uncomfortable conversations,
• 42:23
  we're going to gather here, we'll have a few channels, you can chat and whatever.
• 42:26
  And my experience of seeing those sorts of communities evolve is that you'll start
• 42:32
  out and people will be talking about the episodes and, you know, speculating about who
• 42:35
  your next guest might be or whatever.
• 42:37
  But then there'll be a channel that starts where people want to recommend other podcasts
• 42:40
  to one another and there'll be a channel that starts where people are talking about books
• 42:43
  that they're reading.
• 42:44
  And it becomes the Brooklyn bar, you know, the reason you came there, the reason you move
• 42:48
  to Brooklyn is because you like the podcast, but it becomes a neighborhood.
• 42:53
  Right.
• 42:54
  So the alternative for the Brooklyn bar is the Alabama saloon in the 50s, right, which
• 43:03
  is a white-sonally space or a Nazi bar or a Nazi bar, but let's not even go that far
• 43:09
  because we can get there.
• 43:10
  But I think that can sometimes needlessly inflame things because some people don't like
• 43:16
  that analogy.
• 43:17
  But, you know, just a bar where, which is so like-minded as to be hostile to interlopers,
• 43:25
  right?
• 43:26
  And then it becomes boring.
• 43:27
  Really?
• 43:28
  It becomes closed.
• 43:29
  And it's an echo chamber.
• 43:31
  Yeah, that's right.
• 43:32
  How do you- why is that not what you're creating with the Blue Sky groups, you know, with this
• 43:38
  shift away from the internet being a place of-
• 43:43
  Because I don't think it's homogenous.
• 43:45
  So some of the early, very successful custom feeds on Blue Sky, Clare Bell started one
• 43:51
  called Book Sky, which is sort of a move from book talk, but all the people who are
• 43:55
  interested in book recommendations, authors posting about their new books, you know, that's
• 44:00
  not homogenous.
• 44:01
  There's a great bike sky feed, which is all about people who are into biking, urban
• 44:05
  design around biking, you know, like these are that communities that start out around
• 44:11
  a shared interest, which is how we use to find our Kendra's spirits online.
• 44:18
  Right.
• 44:19
  Aren't they politically self-selected as well?
• 44:21
  I haven't been on Blue Sky, but I know that the initial assumption was that it was-
• 44:27
  You know, I even heard left-wing friends of mine saying, oh, it feels a little bit
• 44:30
  nervous.
• 44:31
  You know, it feels- it does feel a bit like, you know, now the good progressives are back
• 44:36
  in charge.
• 44:37
  We can-
• 44:38
  There's a joke that Blue Sky can't take a joke, but that hasn't been my experience.
• 44:43
  I don't know.
• 44:44
  I guess it is where you, you know, you might- what even able who you weren't up in.
• 44:48
  Yeah, okay.
• 44:49
  You mentioned in passing that AI has ruined search, not everyone will be across that phenomenon,
• 44:55
  but they might have noticed Google degrading.
• 44:58
  But Google doesn't work in it.
• 45:00
  Okay.
• 45:01
  Do you just explain what you mean by that?
• 45:02
  Sure.
• 45:03
  So, I mean, a few things have happened as Google has degraded, I think.
• 45:07
  The first thing we saw was a lot of sponsored content, so you could pay to have your results
• 45:12
  show up in search near the top.
• 45:16
  And we all learned to scroll past the first three sponsored links, and then people would
• 45:23
  train game those sponsored links, so all they really want is the click-throughs for ads
• 45:28
  and so on.
• 45:29
  So, it doesn't really matter what's on the page.
• 45:31
  And a quick way to fill a page is to generate a lot of AI Slop, so, you know, you go into
• 45:39
  chat GPT and say, best hotels in Sydney, write me reviews, and then you slap that on
• 45:45
  a website with the SEO tags, best hotels in Sydney, and then the next time someone searches
• 45:50
  best hotels in Sydney, it's a, or a Boris of AI Slop that's sort of self-fulfilling.
• 45:57
  And now Google is applying its own AI to the top of those search results.
• 46:01
  So, now the first...
• 46:02
  Try to figure out what AI should try to move for quality control?
• 46:06
  No, no.
• 46:07
  So, now when you go search best hotel Sydney, the first bit will be a paragraph from Google's
• 46:12
  AI attempting to extract what it knows, so that you don't even have to go to the search
• 46:17
  results.
• 46:18
  That's going to tell you answer.
• 46:19
  And of course, famously, so far it is recommended that people eat rocks and put glue on pizza
• 46:24
  and, you know, because it's not good.
• 46:29
  No, Gemini is not good.
• 46:32
  Is the problem just that it's not good?
• 46:34
  Like if it was as good, if chat GPT, which, you know, maybe the gold standard correct
• 46:38
  me if I'm wrong, if that gets 10 times as good in the next three years, is there an inherent
• 46:44
  problem with search being...
• 46:46
  Yeah, I mean...
• 46:47
  The first thing is it can't get 10 times as good in the next three years.
• 46:51
  It's run out of training data and now it's eating its own vomit.
• 46:53
  So, it continues to scrape the internet, which is increasingly full of AI-generated content.
• 46:58
  So, there is no room for exponential improvement.
• 47:01
  Well, I think there's no room for exponential increases in input, because room for exponential
• 47:07
  increases in the capacity of the large language models to interpret that.
• 47:12
  I don't believe that there is.
• 47:14
  I think the constraints in terms of computing power and sheer money.
• 47:18
  I mean, OpenAI spent $9 billion last year to lose $5 billion.
• 47:22
  Yeah.
• 47:23
  Like this is not a sustainable model.
• 47:25
  There's no practical use case.
• 47:28
  So, everyone keeps saying, well, it's going to be agents and it's like, there are no agents
• 47:32
  that anyone's prepared to pay for.
• 47:34
  And so, the numbers around revenue and use of goods...
• 47:36
  Not like that.
• 47:37
  Yeah, not like that.
• 47:38
  No, but I'm serious.
• 47:39
  It's like blockchain.
• 47:40
  It's been around 15 years, there's no...
• 47:41
  But blockchain has no purpose.
• 47:43
  No, well...
• 47:44
  But I mean, having, you know, replacing no longer having to pay a convincing lawyer to settle
• 47:50
  a house...
• 47:51
  That's not going to happen.
• 47:52
  What does that mean?
• 47:53
  I say this as a recovering lawyer that is never going to happen.
• 47:56
  And AI is never going to be able to write a contract that's what it's like.
• 48:00
  Why?
• 48:01
  Well, lawyers exist because mistakes get made.
• 48:04
  The smart contracts that have been utilized so far are notorious for getting hacked or
• 48:09
  having errors in them, which mean that they pay out when they shouldn't.
• 48:12
  And that's exactly why lawyers are around because when people write contracts, they misunderstand
• 48:17
  each other or they write down the wrong things or, you know, I mean, as a baby lawyer, the
• 48:23
  very first thing we were told in training was not to mess around with convincing because
• 48:28
  the mistakes you make in convincing are really fatal.
• 48:32
  Yeah.
• 48:33
  And you know, you won't be selling a house when you shouldn't have or buying a house
• 48:37
  when you shouldn't have...
• 48:38
  You seem to be operating in a universe in which human lawyers are infallible.
• 48:42
  No, not at all.
• 48:43
  But they're a lot better than ChatGPT.
• 48:45
  Well, right now they are.
• 48:46
  Yeah.
• 48:47
  But...
• 48:48
  My argument is they will continue to be...
• 48:49
  You think they will always be.
• 48:50
  Yeah.
• 48:51
  I was writing a newsletter a couple of weeks ago and I was writing about the art that
• 48:55
  survives an apocalypse.
• 48:56
  And I was talking about why we always think of Shakespeare as something that survives the
• 48:59
  end of the world.
• 49:00
  And as a gag, I thought I'd ask ChatGPT where the most unusual places were that Romeo and
• 49:08
  Juliet had ever been performed because I wondered whether like it had once been performed
• 49:13
  in Antarctica, Mordo or something like that, like what are the extreme things?
• 49:18
  It straight up made up 30 performances of Romeo and Juliet that have never occurred.
• 49:25
  One in a bond out building in Sarajevo during the war, one in Wales involving live goats,
• 49:32
  and on a Russian submarine where the balcony scene was delivered while Juliet was in a torpedo
• 49:38
  silo.
• 49:39
  Like this was just incredible nonsense.
• 49:43
  Yeah.
• 49:44
  I don't regard any of those anecdotes as persuasive remotely in the sense that anything
• 49:48
  that in any discussion about an audacious future technology, people in the early stages
• 49:56
  of it, who are no sayers, can point to the limitations of it in its current form which
• 50:01
  tells us nothing about why do you think we're in the early stages of it?
• 50:05
  Where do you see it going?
• 50:06
  Given that it's already sucked up all of the internet, essentially, like it's run out
• 50:10
  of input.
• 50:11
  Yeah, but it doesn't just, I mean, when you use it, it's not just spitting out the data
• 50:15
  that it learned from the internet, it's making connections about things and mimicking intelligence.
• 50:21
  It's not mimicking your charges.
• 50:22
  It's spicy autocomplete.
• 50:24
  It's literally looking at what you've put in and predicting the probability of what should
• 50:29
  go next.
• 50:30
  It's just not doing, that is not, I understand that argument, I've heard it a million times.
• 50:35
  That's not the experience that people who use church epiti every day have of it.
• 50:40
  It's not the experience that people who have invested a lot of money and wanting you to
• 50:44
  think that have had of it.
• 50:46
  Well, do you use it on a daily basis?
• 50:47
  Absolutely not.
• 50:48
  Okay.
• 50:49
  Because I mean, it can do a certain limitless amount of planning, lateral thinking, creative
• 50:55
  script writing, analysis of all kinds of things that it can do, that even today, setting
• 51:05
  the use of lateral thinking.
• 51:07
  It's not thinking.
• 51:08
  Well, that's the extraordinary thing about it.
• 51:10
  Of course, it's not thinking, I mean, but it is doing something that if I didn't tell
• 51:19
  you that it was an AI doing it, and I set up a test of like you had to predict whether
• 51:25
  or not this is a living person or not, it's passing the chewing test with flying colors
• 51:30
  on a number of different metrics.
• 51:32
  I think so.
• 51:33
  It's lateral thinking like behavior that I think will, oh look, I have no idea where it
• 51:38
  will go.
• 51:39
  You might be right, and of course there's all these problems of hallucinations, there's
• 51:42
  a lot of problems of how bad Gemini currently is, it's an environmental calamity, you
• 51:46
  know, it's like, it's a complete, but I mean, I regard humanity as a baby giraffe at
• 51:53
  the moment, with covered in goo and trying to stand up and figure out what to do with
• 51:57
  the information explosion that it is living through, like our civilization, the way that
• 52:02
  it talks to itself is undergoing a radical transformation starting with technology in
• 52:08
  the computer and the internet, and now I see AI as, at least as pivotal as the internet
• 52:13
  and probably as pivotal as the discovery of fire or democracy in, and being something
• 52:20
  that we're going to have to grapple with, and that may, in 10 years time, we may just
• 52:25
  look back on my saying that and think that was just ridiculously naive, it might just
• 52:28
  be like, why 2k, totally, it's totally possible, but if the capacity of artificial systems
• 52:36
  to integrate information can continue to increase in the absence of new data being fed
• 52:42
  into them, or if they can get new data, for example, through cameras of their own or
• 52:46
  through direct interaction with the world, just like animals do or plants do, we don't
• 52:52
  have to learn everything we learn it on the fly, then I don't see any reason in principle
• 53:00
  why human level intelligence or chimp level intelligence or something, is there magical
• 53:08
  pixie dust about the human brain that makes it the only object in the universe that is
• 53:13
  capable of sophisticated thinking and planning?
• 53:17
  Yeah, well, I guess we'll find out, I mean, I fundamentally think those two assumptions
• 53:21
  like assuming that it can get exponentially better, and we're already seeing a plateau
• 53:27
  in terms of improvements between chipt-t models, where is it rolling out?
• 53:32
  Which you would always expect, I mean, any, any exponential curve, plateaus at a time.
• 53:36
  Yeah, but your assumption is that it's going to be 10 times better in three years.
• 53:40
  Oh, I mean, that was a hypothetical, that's not my prediction.
• 53:43
  I do think it'll get 10 times better, but it might not be as quickly as three years.
• 53:47
  I mean, I feel like, how does it get 10 times better if it's doing this though?
• 53:50
  If the rate of increases going from 1,000X a month to 500X a month, it's still pretty
• 53:58
  good.
• 53:59
  So there's that assumption, and then there's the assumption that that growth can continue
• 54:05
  because it's possible for there to be more data going into it than there is now, and
• 54:09
  I don't think either of those assumptions are.
• 54:11
  No, to me, something feels off.
• 54:13
  It's a bit like we're in the era of the right brothers, and we're talking about how the
• 54:18
  cloth wings aren't very good.
• 54:21
  And there's not enough.
• 54:22
  We'll get back together in ten years.
• 54:23
  There's not enough cloth in the world to build a bit.
• 54:26
  You know, there's nothing else going into the aeroplane.
• 54:28
  There's nothing else going into planes.
• 54:30
  Anyway, so that just got us off the topic of search and Gemini and AI and so on.
• 54:37
  But what that's meant for search is that it's very difficult to discover things now.
• 54:42
  Right.
• 54:43
  And in the early internet, it was also hard to discover things.
• 54:46
  We didn't have search engines in the start, and we relied on hand-curated directories
• 54:53
  of links put together by individuals, and in fact, Genama Polly, who's the librarian who
• 54:59
  coined the phrase, surfing the internet, she said the reason that she liked that metaphor
• 55:04
  is because, like, the sport, it required skill and patience and a sense of adventure.
• 55:08
  Like, it wasn't easy to get around.
• 55:11
  And so I'm really interested that we've sort of gone full circle on making things very
• 55:16
  easy to find, and then back to making them very hard to find again.
• 55:21
  And so what does discovery look like in that new environment?
• 55:24
  And, you know, I don't think it's the case that we go back to web rings where we sort
• 55:29
  of link to each other's web sites.
• 55:30
  What was that?
• 55:31
  I was, like, early websites would often have a sort of onto the next one arrow on the
• 55:36
  bottom, and you sort of have a ring of websites that you could flip around.
• 55:40
  But how do we think about discovery?
• 55:42
  So, you know, Kevin Rosen, Alexis Ahanian, and bringing back Dig, which used to be sort
• 55:48
  of one of the ways that you found new things.
• 55:51
  But what does a genuinely new system of discovery look like?
• 55:55
  Like how do we go exploring for new and interesting ideas and people and topics and communities
• 56:01
  in an era where search is increasing?
• 56:05
  Well, what's the answer to that?
• 56:06
  Is it human curation?
• 56:08
  Could be.
• 56:08
  I mean, one of the things I've been looking at lately is a concept of thinking more about digital
• 56:14
  gardens, which is a phrase that some people use.
• 56:18
  So, a lot of what we have done over the last few years is think about the Internet as a broadcast
• 56:23
  medium, like it's, if you're going to publish something that's polished and finished and
• 56:27
  pushed out there as a podcast or newsletter or so on.
• 56:31
  But what does it look like to share much more of just the ideas you're interested in,
• 56:35
  the things you're thinking about and think of that as like a garden, like some of
• 56:39
  them are very untended ideas, some of them are, you know, might grow into things.
• 56:43
  And how do we encourage people to explore those gardens?
• 56:46
  And we use the garden metaphor negatively at the moment, like walls gardens, we talk about.
• 56:51
  But what does it look like to take down those walls and have people be able to wander again?
• 56:56
  And, you know, I'm fascinated by that idea.
• 56:59
  And I think some of it will be tool driven, like Blue Sky or, you know, there's a new
• 57:07
  app called Tapestry, which brings all of your feeds into a single place.
• 57:12
  I don't know if I want that.
• 57:13
  Like, I don't know if I want to be scrolling my hockey tumbler thirst posts alongside US political
• 57:19
  doom. But I do think the idea of thinking about new ways to exercise sort of
• 57:26
  agency over our journey online is going to be really crucial in the next.
• 57:30
  Can you think, so what would that look like?
• 57:31
  Are you able to put flesh on the bones of what that would feel like just to articulate what we
• 57:36
  mean by a garden? Like, so a lot of people who, so I will give you an example.
• 57:43
  I've started writing this newsletter this year.
• 57:45
  It's once a week and it's a collection of things that I am thinking about, right?
• 57:49
  So that's sort of my mental garden and it's links to articles I've read, things that I've heard
• 57:54
  online, things that I'm kind of thinking about. But I think the idea of a digital garden would be
• 58:01
  having that on my site, so within my control and the sense of digital permanence,
• 58:05
  but in a way that other people can explore and draw their own connections through, right?
• 58:09
  Because at the moment, unless I say in the newsletter, oh, two weeks ago when we were talking about
• 58:14
  the Conclave movie, you know, and now we're talking about the death of the Pope,
• 58:18
  I have to draw that link for my audience for them to find that connection.
• 58:23
  But if all of those things were on my site, if my site was linked to other things,
• 58:27
  if there was bidirectional linking the way we used to do in the old days,
• 58:30
  people would be able to sort of move from reading my piece about Conclave fandom to Adrian
• 58:38
  Hans piece about Conclave laps to another article about finished sci-fi laps, and you're able to
• 58:45
  wander. So is this an interface that doesn't yet exist?
• 58:48
  I think there's a lot of people trying to tackle it, so there's an app that's just gone public
• 58:53
  in the last two weeks called Sublime and their thinking is related ideas. And so it's kind of a
• 59:00
  bookmarking tool while you're surfing around, you see something and you're like, oh, that's
• 59:04
  interesting, and there's an extension and a nap and so on, so it makes it easy for you to bookmark
• 59:09
  that idea, and you chuck it in with your own tags, like we're back to the web.2 error of folks
• 59:15
  on them is, and so I might say, oh, this is interesting for my newsletter and thinking about this
• 59:19
  topic, but the app will then show you related ideas, and it's learning that based off the content
• 59:26
  in the excerpt that you've clicked with the site that you've read, and based on how other people
• 59:33
  have tagged it and arranged it and so on, and so instead of just seeing your bookmark,
• 59:38
  you're then able to go, oh, yeah, that's exactly the point, and click through to something else.
• 59:44
  Isn't part of the pitch of, is it notebook AI, or one of them, that it can be, or maybe it's
• 59:50
  an app like Notion or something, and I'm getting confused, that you can just dump everything in there,
• 59:55
  and it's a space for your brain, and then it'll make intuitive connections between things. So you never,
• 1:00:00
  I mean, look, it's already happened in things like, I was remembering that I used to have
• 1:00:05
  Gmail folders that I would put particular emails into, so that if I needed to find something,
• 1:00:10
  I could find them again, right? You know, like, tax and banking information goes in this particular
• 1:00:15
  email folder, who has email folders anymore? The search function is so good that no matter what I
• 1:00:20
  need, that's fine. It was so good within your Gmail, isn't it still? I don't think that's changed.
• 1:00:26
  Gmail's okay, Google drivers amiss, but yeah, yeah, yeah, yeah. Why is Google drive that time?
• 1:00:31
  No, it's extraordinary. It throws up documents that have nothing to do with the search time.
• 1:00:35
  You will never find a document again, like it's just strange. It's actually safer for me to not
• 1:00:40
  save attachments in my Gmail to Google Drive, because I'll be able to find them in Gmail,
• 1:00:45
  but they don't seem to use the same self-functionally over in Drive. Anyway, hopefully,
• 1:00:50
  even if AI gets no better than it currently does, it should be able to assist
• 1:00:56
  better than it does at the moment in searching for things. So is there a way of corralling all of
• 1:01:01
  the information and your interests and your friendship groups and so on into a, I mean, is this what
• 1:01:05
  you're basically envisaging? Well, it's sort of an app or a platform. It's certainly what I'm
• 1:01:10
  thinking about this very week. This is what I was just writing my newsletter before I came here,
• 1:01:13
  and this is what I was thinking about, which is these concepts aren't new. So the Web 2.0
• 1:01:19
  era was when we first saw the idea of tags tagging, and Fox Onamies, and this was the idea that instead
• 1:01:26
  of having a top-down text on me, like your email folders, banking and text goes in this folder,
• 1:01:32
  you could just tag anything, banking, tags, to pay, invoice, whatever, and there you were saying
• 1:01:39
  to pay like a hair piece. So I was like, why does she think I buy hair pieces? What a fine
• 1:01:44
  head of hair. So you didn't then have to remember necessarily where it was because the tags would
• 1:01:52
  enable you to do it, and you could do community tagging, right? And so the early bookmarking sites
• 1:01:57
  like Delicious and Pinboard were all about shared tags because of I tag something, you know,
• 1:02:03
  fanfic, everyone can see that that's an article that's relevant to that topic. And I'm someone who
• 1:02:12
  always falls victim to the idea that I'm going to be great at some sort of bookmarking,
• 1:02:15
  organizational, I just need the perfect notebook or the perfect pens. It's not going to happen.
• 1:02:20
  And so, you know, I tried Delicious for a while, I tried Evernote for a while, which was another
• 1:02:26
  one that was like... I was a big Evernote user, yeah. Your whole brain in there, yeah. I don't know
• 1:02:31
  what I saved there, but it's gone now, you know, so... What happened? Did Evernote get bought or
• 1:02:35
  something, or did we just drift away from it? Both, I think. I don't know. There'll be some
• 1:02:40
  passionate Evernote user out there who's like, oh, there's probably all kinds of... Yeah, that's true,
• 1:02:44
  yeah, there's probably all kinds of stuff like that. So, you know, these tools aren't, they're not new.
• 1:02:49
  And I do think there's something to be said that that no note-taking app or notebook or whatever
• 1:02:55
  is the answer to thinking, you know, like it doesn't matter if a computer draws all these connections
• 1:03:00
  for you. There's no replacement for going for a walk in the park and then sitting down and actually
• 1:03:06
  trying to draw some conclusions yourself. Do you still do that enough? I was in the park just
• 1:03:11
  before I came here. Lovely. Does that take a conscious effort or are you just drawn to it?
• 1:03:17
  No, I think it does take a conscious effort. I had a 13 and a half hour zoom call yesterday,
• 1:03:21
  so I needed to see some trees. Well, I do worry about that as well. The fact that we're always on.
• 1:03:30
  Like, there's never a break. People are going crazy. People are not getting the mental space
• 1:03:37
  just as on the side, by the way, when you were talking about you're always hunting for the next
• 1:03:41
  life hack that's going to solve everything and fix all of your to-do's. Do you know Oliver Berkman,
• 1:03:46
  the writer? I'd strongly recommend his books and he's been on this show a bunch of times and people
• 1:03:52
  who are listening to this who haven't heard those episodes should go back and listen to
• 1:03:55
  Oliver Berkman. He's a Brit who writes about how you're never going to get to the end of your
• 1:04:01
  to-do list, focusing on it, worrying about the life hacks and find a way to
• 1:04:06
  befriend your own existence in the space of chaos and sort of optimistic resignation.
• 1:04:12
  Yeah. Quite lovely. That sounds like me. I'm never not injecting my consciousness with
• 1:04:21
  data of some kind, whether it's a podcast in my ear or whatever. The mental space that I have
• 1:04:28
  to reflect is when I'm in the shower, basically, or when I'm reading fiction, which is not exactly
• 1:04:34
  an empty space either. No. I like many people know that I ought to meditate. I know that I
• 1:04:39
  ought to walk in the park more, right? I live at a beach and it's a challenge for me to force myself
• 1:04:45
  to go down and walk along it. That should be a rule. That should just be every morning,
• 1:04:51
  first thing I do is I walk along the beach and I come back. But it's true that you don't come up
• 1:04:56
  with the good ideas or the good connections until you give your brains some quiet and this year
• 1:05:02
  I've been trying to implement for myself a bit more of a digital Sabbath and just getting off my
• 1:05:08
  devices for one day. Yeah. And how's that going? It's been really interesting. I start the evening
• 1:05:15
  before like a religious Sabbath because there's something very satisfying about having a quiet evening
• 1:05:21
  and going to bed and knowing that you're going to wake up to a quiet morning. But also in the morning,
• 1:05:25
  you wake up and you think, well, I can get back online but at the end of the day.
• 1:05:30
  And so that sort of frees me up. I'm like, I don't feel like my throat's been killed. I was
• 1:05:33
  not like it ends at sunset in the second day of the Sabbath. And I bought myself a dumb phone
• 1:05:41
  because that alleviated. I found the first few times I did it. I was constantly like, what if
• 1:05:45
  there's an emergency? What if someone's trying to get hold of me? What the dumb phone? I've been
• 1:05:49
  looking. I'm not sure if you're going to want it but it is a Barbie phone. It's bright pink.
• 1:05:57
  But it's made by HMD which is the crowd who I think have bought the Nokia IP and so they're
• 1:06:02
  turning out the old Nokia's again. So you can buy like a Nokia brick or whatever.
• 1:06:06
  Right. And can you send text messages on it? Yeah. And you do have to tap it three times to
• 1:06:11
  write. And I really thought that that might have been muscle memory, that T9, like I,
• 1:06:15
  because I used to be fast. Yeah, me too. But it turns out no. No, you've got to re-learn it.
• 1:06:20
  Got to re-learn that. So that helped a little bit with the noise of my brain like,
• 1:06:24
  oh, someone will be trying to get hold of me. It's like, no, they can call or text. That's fine.
• 1:06:29
  Is it the same phone number though? Oh yeah, I just put the simon.
• 1:06:32
  Right. It's been annoying.
• 1:06:36
  One little friction. You're already out. I'm done. I'm gone. No, I, yeah.
• 1:06:43
  And so, so then it's just off my devices. And by devices, does that include television?
• 1:06:49
  Yeah. I read books. Sometimes I'll let myself listen to an audio book but I play it through a
• 1:06:55
  speaker so I'm not carrying the phone around. But yeah, just one day. And what I'm finding is
• 1:07:02
  that is the time when the connections are starting to be made. I mean, that's when I'm like,
• 1:07:09
  that article feeds into this or I should seek out a podcast about that or, you know,
• 1:07:14
  we all know we should do that. We just don't do it. So Nazis. Nazis in the bar. What's the
• 1:07:21
  analogy about tell us what you mean by Nazis? It's a, it's a quasi famous Twitter thread but it was
• 1:07:26
  somebody who was saying that the problem with allowing or platforming people on social media who
• 1:07:37
  are expressing really offensive points of view is that if you're a bartender or bar owner and
• 1:07:44
  you let one person in who's sitting at the bar not causing any trouble but they're wearing
• 1:07:49
  Nazi memorabilia, they've got their tattoos, whatever. If you don't immediately kick them out,
• 1:07:55
  then the next time they'll come and they'll bring a friend and then Sponda, you look up and turns
• 1:08:00
  out that you're running a Nazi bar. And it's a, it's a sort of, it's a simplistic way of thinking
• 1:08:04
  about platform neutrality and I think that one of the characteristics of the Silicon Valley mindset
• 1:08:12
  over the last couple of years has been that the platforms not publishers are not in the business
• 1:08:18
  of making editorial decisions and that, that therefore they shouldn't be taking steps to
• 1:08:28
  sensor or take steps around content. And I think the first few years of the social media era,
• 1:08:36
  we saw the limitations of that. Amnesty International did some work that said that for a woman
• 1:08:41
  of color, they received an abusive authoretning to eat every 20 seconds. And so it was very clear
• 1:08:48
  that it was becoming an unsafe space as it grew. And so most of the platforms invested time in
• 1:08:55
  resources and trust and safety teams in trying to work out how to ensure that their platforms
• 1:09:04
  encouraged conversation but didn't wind up in these sort of very negative outcomes. And then,
• 1:09:11
  of course, you get this swing in the Trump era where the argument becomes, these platforms are
• 1:09:19
  censoring particularly alt-right voices, you know, that they're being shut down and silenced.
• 1:09:25
  And the, the data shows that that's not true, that actually when you look at the work that was
• 1:09:30
  done by the trust and safety teams in that phase, that right-wing voices were not affected at any
• 1:09:37
  greater rate than left-wing voices. But nevertheless, that was the, that was the sort of sentiment that's
• 1:09:43
  stuck particularly in the United States. And so that's how you then get people like Elon Musk who
• 1:09:47
  are saying, I'm going to buy Twitter and I'm going to make it a haven for free speech. And what he
• 1:09:52
  means by free speech is the ability to use slurs again. And then you get Zuckerberg coming out
• 1:09:58
  shortly thereafter saying that he'll also be dismantling trust and safety at meta because he'd
• 1:10:04
  like to be able to, you know, use the R word. And so, so with now come sort of down the other
• 1:10:11
  side of the hill, I guess, that there isn't, there isn't teams looking at these things on the
• 1:10:16
  platforms. And that's why a lot of people have left Twitter and why some people have stopped using
• 1:10:23
  meta products. I mean, it's a big, it's, this is a biggie, right? Like the question of how
• 1:10:31
  how big tech should function in a chaotic environment that is also democratic and that also
• 1:10:40
  prioritizes traditional small L liberal ideas about free speech in the public square,
• 1:10:48
  crashes into the algorithms, right? I mean, let's just put a few things on the table. The existence
• 1:10:56
  of the algorithm means that social media companies can't plausibly claim to not be publishers.
• 1:11:01
  Correct. Right. Because they are publishing, they are choosing what goes at the top and what
• 1:11:04
  doesn't get seen. Yeah. So they're making publishing decisions. They're not just platforms. If they
• 1:11:09
  want to be treated as platforms where they have no culpability for what gets said on them,
• 1:11:14
  then they can remove the algorithms and turn your feed back into a reverse chronology of the
• 1:11:18
  people who you follow. Yeah. And problem solved. So that's bullshit. It's also bullshit for
• 1:11:25
  Elon to bang on about free speech at the same time as he sues people, sues advertise,
• 1:11:31
  sues companies for not advertising on Twitter anymore, compelling them to, you know,
• 1:11:34
  get that engage in speech that they don't want to engage in. What is the people that you mean about
• 1:11:39
  Tesla? Yeah. Exactly. Kicks people off, you know, Twitter for criticizing him, you know,
• 1:11:47
  gets butt hurt when people publish, you know, details that he doesn't find appropriate.
• 1:11:54
  I also think, people sometimes forget that these are private companies that are allowed to make
• 1:11:59
  their own decisions about who says what? They're not the government. This is not a first amendment
• 1:12:04
  like question in the United States, right? Nobody has a right to say anything on any particular
• 1:12:08
  platform. They could easily make the platforms super, super nice. They could easily make the
• 1:12:15
  platforms super super Nazi. Yeah. They could easily make them super lucky. They could make them
• 1:12:19
  super right. Yeah. That's right. Gab or parlor or that's right. Yeah. And from my perspective,
• 1:12:25
  if you want to start up a neo Nazi social media site, then that should be legal. I think you're
• 1:12:31
  I know as long as you're not actually advocating for harming people, you're allowed to throw around
• 1:12:35
  whatever bullshit ideas you want to in private. The question is whether or not you should do that on
• 1:12:40
  social media platform that all of us want to use and that has basically become a kind of a utility
• 1:12:45
  essentially for public conversation. That all being said, the question of whether social media was
• 1:12:53
  doing a decent good enough job prior to Elon. Oh, absolutely. I was allowing. No, no, not at moderating.
• 1:13:00
  Oh, I see. But at being even handed and not getting caught up in its own echo chambers,
• 1:13:05
  I think is still an open question. Like the, you know, there are many famous cases which are not
• 1:13:12
  apocryphal about people suggesting that lockdowns were going too far, for example, whose
• 1:13:20
  feeds were suppressed or tweets were mysteriously never got any views in comparison to the rest of
• 1:13:28
  their content. Sure. I mean, there's we can put it off the platform for saying men aren't women
• 1:13:32
  because trans ideology got, you know, quite extreme at that point in time. That's true. That's true.
• 1:13:37
  That is true. You know, the lab leak theory, which now is probably a 50 50 shot and you know,
• 1:13:43
  scientists aren't quite sure about, but several big American governmental bodies think is more
• 1:13:48
  more likely than not was initially separate. I think we can I think we can pick and choose
• 1:13:52
  examples on both ends of the spectrum. What I'm saying is that there is. But can you on both ends of
• 1:13:57
  this spectrum? I don't think. And there is research and I would need my phone to find it for you,
• 1:14:02
  but there is research that shows that moderation decisions were affecting all ends of the political
• 1:14:09
  spectrum. That right wing, alt right, whatever labels you want to apply, voices were not being
• 1:14:17
  targeted and were not subject to moderation decisions at any greater rate than any other political.
• 1:14:23
  I mean, it would be remarkable if that were true since all of my cohort of people who were
• 1:14:30
  the types of people who run Twitter were on the left. And certainly at that time coded
• 1:14:38
  the right as being racist. But during that era, did you experience people on the left
• 1:14:43
  feeling like the moderation decisions were going well? Or did you feel that, you know,
• 1:14:49
  left wing voices were saying this is outrageous. Why is this still here?
• 1:14:55
  Both. I mean, I feel like the let I feel like we're talking about peak. It didn't turn into a
• 1:15:01
  lefty echo chamber. Well, I mean, you've got several ongoing forces which are sort of in
• 1:15:09
  confluence, right? So there's the there's the rise of the chaoticness of the of social media,
• 1:15:15
  which is pushing back against any sort of social justice hegemony that the corporations might
• 1:15:22
  have been imposing. So it's possible that it's possible that a couple of things happen at once.
• 1:15:27
  Jack Dussi was imposing a social justice hegemony or Mark Zuckerberg. Well, I got a chance.
• 1:15:34
  Yeah, I mean, yeah, no, yes, I do. I think that's a fair thing to say because at the time in
• 1:15:39
  between 2015 and 2022, there was a lot of work tweets online. No, there were lots of things online,
• 1:15:49
  but the corporations were extremely mindful of getting themselves into shitstorms
• 1:15:56
  and were extremely mindful of at least taking seriously or paying lip service to the need to
• 1:16:01
  quote unquote stop the spread of hate or hateful ideas or whatever. And a lot was being coded as
• 1:16:07
  hate that people on the right regarded as being a legitimate part of public discourse. For example,
• 1:16:15
  you know, decreasing the rate of immigration or whether or not, yeah, the vaccines were doing
• 1:16:20
  damage or whatever it might be. Like I do think I don't think I don't think you can
• 1:16:25
  I don't think it's tenable to say that there isn't a conflict about where exactly the parameters
• 1:16:29
  of the conversation lie on social media that was during that period filtered through the prism
• 1:16:38
  of a particular political and social justice worldview. That's what Elon was claiming to be pushing
• 1:16:44
  back against. And he's now just growing up. Yeah, and I'm saying I'm saying statistically
• 1:16:49
  that wasn't the case in terms of I suppose it depends how you count what voices like what constitutes
• 1:16:55
  the right while constitutes the left. What opinions constitute which who did that study? Yeah, yeah.
• 1:17:00
  But I mean, but Trump and Elon like to bang on about the fact that right wing voices were shadow
• 1:17:04
  banned and silenced and and taken down and and whatever. And the the research that has been done
• 1:17:11
  says that that's not the case. Right. And and you know, I think if we're picking and choosing
• 1:17:18
  examples, there are plenty of people on the left who would say that there wasn't nearly enough
• 1:17:23
  done in that era to stop the spread of hate. And you know, when we've got people like Graham
• 1:17:30
  Linnahinn and so on going on and on on Twitter, finding followers in a platform and a home there.
• 1:17:38
  So it's the idea that right wing voices were censored and left wing voices created a little
• 1:17:45
  you know, happy neighborhood for themselves where they didn't have to hear any
• 1:17:48
  country reviews. It's just nonsense. Right. That wasn't what it was like. And to people who don't
• 1:17:51
  know Graham was a UK is a UK comedy television comedy writer who came very obsessed with the
• 1:17:57
  transition became a turf for one to be better word. Yeah. Okay. That's interesting. I mean, I
• 1:18:04
  I guess I'm somewhat sympathetic. I'm somewhat sympathetic to the point of view that says that
• 1:18:11
  moderation is always going to be a futile game of whack-a-mole.
• 1:18:18
  Moneration is hard and I am not here saying that it's easy, but I'm saying that it is possible. So
• 1:18:23
  for example, we know on social media that all of the platforms have to comply with German
• 1:18:28
  legislation around what can and can't be said about Nazis online. And so you know, they have
• 1:18:37
  always been able to take content down to meet regulatory requirements. They just choose not to.
• 1:18:44
  Right. But the regulation has to be clear. I mean, the good thing about the German regulation
• 1:18:48
  is it's like, this is what you can say. This is what you can't say. End of story. It's a lot
• 1:18:53
  more difficult to understand. When is the thing that you're saying about pediatric gender,
• 1:19:02
  care, bigotry towards trans people and when is it a defensive feminism? Like these are just
• 1:19:07
  calculations that Mark Zuckerberg is not equipped to make. So he urs now on the side of saying,
• 1:19:13
  all right, we're going to wipe our hands off it. And yeah, and it's going to be more of a free
• 1:19:16
  for all. I think it's unfair to say that the reason why he changes that is because he wants to
• 1:19:20
  say the R word. I think it's because I mean, there are obviously cynical corporate reasons for it.
• 1:19:26
  And look, maybe he's a secret racist who wants to, you know, say terrible bigotry things.
• 1:19:30
  But I think it's more likely that in addition to carrying favor with the Trump administration and
• 1:19:34
  with the right, he just doesn't think that it's consistent with like the long term future of a
• 1:19:42
  a place that is predicated on having conversations to micromanage those conversations.
• 1:19:47
  I just think he thinks that it's hard and too expensive. It's probably a bit of that too.
• 1:19:53
  It's definitely possible. But it involves investing and trust and safety teams that it involves,
• 1:19:58
  you know, human moderators. I mean, one of the things that was super interesting
• 1:20:03
  in Brazil when TikTok was briefly banned on a telegram, sorry, was briefly banned in Brazil
• 1:20:11
  because of the spread of misinformation. And in order to get it back, what they did was they focused
• 1:20:16
  on a tiny handful of super spreaders. And what they discovered was if you moderate this very small
• 1:20:24
  number of accounts who are responsible for amplifying the vast amount of, in that case,
• 1:20:31
  medical disinformation, you can reduce it by an enormous amount. So it's not the case that you
• 1:20:38
  have to have a football field of human moderators who are individually reviewing each tweet and
• 1:20:44
  going is on the side of the line or is on that side of the line. But you have to care about
• 1:20:48
  creating a healthy, vibrant space for conversation. And part of that is making sure that people
• 1:20:55
  aren't being harassed and mistreated and that flat out lies aren't being passed around about
• 1:21:02
  with an election with Stalin. I mean, I think it's clear that if there's that you could draw a
• 1:21:07
  line in the sand for abuse and threats, right? The thing that I honestly find most personally
• 1:21:13
  distasteful about Twitter whenever I'm there is, and again, now, even now that it's a dumpster
• 1:21:21
  fire and I don't even want to go near it, it's not just that there's lots of, it's not that there's
• 1:21:26
  an over-abuse and threats, although as I said before, that that might be different if I was a
• 1:21:30
  woman of color, for example. But it's that the reaction to ideas that people disagree with
• 1:21:37
  is so hostile and so unsympathetic and so instantly disagreeable that the tone is completely off.
• 1:21:48
  And it strikes me that if you actually wanted to create a conducive place where normal
• 1:21:53
  functioning emotionally well-regulated human beings want to converse, you'd have to address
• 1:21:58
  that, which is a lot trickier than getting rid of full-on outright misinformation. Like in some
• 1:22:07
  ways, I sometimes feel like this conversation, especially when conducted with policy bureaucrats
• 1:22:12
  or governments who are looking for simple solutions, it can collapse into a conversation about
• 1:22:18
  misinformation is bad, therefore let's get rid of misinformation, abuse and threats and hate
• 1:22:22
  speech are bad, therefore let's get rid of that. And it strikes me that that leaves this
• 1:22:26
  yearning universe of gray cases that are where the actual problem lies, which is that it's not
• 1:22:34
  misinformation, it's information that is skewed through a political agenda or a prism that
• 1:22:40
  is relentlessly focused on one side and more than the other. And it's not abuse or threats
• 1:22:46
  per se, they're easy to catch and how I could catch those. It's a tone of hostility and antagonism
• 1:22:54
  and close-mindedness towards and tribalism and antagonism and a defensive crouch about
• 1:23:01
  one's own ideas. And this is, I keep looping back round to scale, but I think that's the point,
• 1:23:06
  right? That is the consequence of the scale at which these platforms are now operating. Like there's
• 1:23:15
  I don't know if you're familiar with Dumbars number, but it's the number of stable social
• 1:23:18
  relationships that we can all have as people, right? And it's 150. So the theory is that you
• 1:23:26
  you can have 150 people in your life that you have an ongoing social connection with. And those
• 1:23:33
  sort of layer out close friends and neighbors and so on, but wider, but any more than 150 and it's
• 1:23:39
  it collapses. And I think there's something in that that it's not a that's something like Twitter
• 1:23:46
  where we're now millions and millions of users shoved into one space with an algorithm that's
• 1:23:54
  trying to genus all up against each other. That's never going to produce a healthy conversation.
• 1:23:59
  It just can't. So last question about kicking Nazis out of the bar.
• 1:24:09
  The sort of liberative libertarian in me thinks it's all very well if we get to define who the
• 1:24:17
  Nazis are and where and control of the platforms. What if President JD Vance is the one who's
• 1:24:26
  determining what is hateful speech and what is corrosive speech and what if something like the
• 1:24:35
  Polish right wing or Victor Orban were in power. I think we're seeing the start of that in the US
• 1:24:46
  that's right. And so then the definition of who forms the what jigsaw piece fits the
• 1:24:54
  hole in the mosaic that you're calling a Nazi could just as easily be depraved people like me who
• 1:25:01
  are in dysfunctional marriages with members of the same sex and thereby contaminating traditional
• 1:25:05
  family values or migrants as we're seeing now. My dad was a refugee you know people who you know
• 1:25:14
  the way that they're treating people just pulling them off the streets and throwing them into
• 1:25:17
  gulags in Central America. My gut says like be as careful as you possibly can about
• 1:25:28
  trying to define a way or exclude ideas that you regard as being reprehensible because you
• 1:25:34
  don't want to set a precedent that the next person who disagrees with you will do the same.
• 1:25:39
  Yeah 100% and you know one of my best friends in London is the grandson of a Holocaust survivor
• 1:25:47
  and who used to take him to the soapbox in Hyde Park and and get him to watch people saying
• 1:25:53
  extremely unpalatable things because his view was exactly that you know like that that starting
• 1:25:58
  to decide what is and isn't good speech is a slip or slope to you know. Unfortunately I think what
• 1:26:05
  we're seeing in the US though is is sort of the opposite of that it's like this is the natural
• 1:26:11
  endpoint of platforming some really dangerous ideas over the last 10 years and I think that now
• 1:26:20
  we have to be really careful about what happens next but I do think there's a distinction there
• 1:26:24
  like as you were saying earlier social media platforms are private companies and that's different
• 1:26:31
  from what a government is saying about what is legally protected speech or not in the case of the US
• 1:26:39
  and so when we're thinking about the health of online conversations we're talking about
• 1:26:44
  what those that social infrastructure that companies build that we can build
• 1:26:49
  looks like and how we can make that places that people want to be as opposed to
• 1:26:54
  you know nightmare dumpster fires without they're getting harassed and the number of people you
• 1:27:02
  meet now who sort of say oh I'm just off social media it's so toxic yeah and it's like that's fine
• 1:27:07
  that's a perfectly plausible decision to have reached but I think we're losing something we're
• 1:27:11
  losing something that was at the heart of the early internet that we are in a nostalgic for which
• 1:27:17
  is that community piece that that ability to find people who love the same things that you do
• 1:27:24
  that are interested in conversations about fascinating topics who are going to expose you to new
• 1:27:29
  ideas and and give you new things to think about that doesn't have to be social media though does it
• 1:27:34
  no yeah it's not because I'm off social media basically I mean I I use it as a publishing platform
• 1:27:38
  to disseminate you know clips of this show but I don't handle that is there anywhere that you
• 1:27:45
  hang out online now well interestingly yeah yeah so I was going to I don't spend a lot of I
• 1:27:54
  am intentionally minimizing the amount of time that I spend online in all capacities in order
• 1:27:59
  to touch the grass more as they say but the main thing that I'm optimistic about would be
• 1:28:03
  sub-stack where she rolls her eyes because as long as it doesn't as long as it continues to
• 1:28:12
  not deploy algorithms I think it is a place where you can curate your experience in a way that
• 1:28:18
  feels a bit more like do you not think it's already deploying algorithms um well it's it's not
• 1:28:23
  presenting me with anything that I'm not uh actively proactively following so the experience
• 1:28:30
  put anything in front of me yeah so that the experience for someone signing up to follow a sub-stack
• 1:28:36
  now is that the very first thing like you've just come across someone you're like that guys
• 1:28:41
  it's interesting and the photo has sub-stack the first thing that happens is it will insist that
• 1:28:45
  you go across to the app so it is no longer the case that you just get to sign up for a newsletter
• 1:28:51
  and receive a newsletter in your email the next thing it will do well I think you can do it on
• 1:28:56
  laptop unlike desktop you mean if you're on your phone it'll it'll it'll automatically
• 1:29:02
  the sub-stack it yep okay then it will show you three other newsletters that it thinks that you
• 1:29:07
  should also subscribe through so you can either choose to do that or skip it like there's some
• 1:29:12
  dark patterns around how to actually get to what you wanted which was just getting a newsletter
• 1:29:16
  from you once a week and what that's then doing is juicing the growth numbers for people because
• 1:29:22
  they get a whole influx of new free tier subscribers based off this sort of sign up pattern which is
• 1:29:28
  all algorithmically driven oh it's not all algorithmic because the use the publisher can and I
• 1:29:32
  know because I'm a publisher on sub-stack as well can choose who they want to yeah so if you don't
• 1:29:38
  it's actually still going to show your users content yeah and what that's doing is creating
• 1:29:47
  growth numbers that they can then point to for new writers and sort of say look this is a growth
• 1:29:51
  engine you can get all these sign ups but of course those ones churn and if you have a paid tier
• 1:29:55
  they don't convert at the same rates so it's a growth hack it depends what your public it depends
• 1:30:00
  whether you're good mind on chan it's a it's a growth well okay it's a growth hacking initiative
• 1:30:07
  from sub-stack the other thing I struggle with with them is and I've been very public about this
• 1:30:14
  is that they are fully aware that they have a Nazi problem and have been very public that they
• 1:30:20
  are not intending to do anything about that and so that means that if you are a paid author on
• 1:30:28
  sub-stack you are funding the spread of Nazi content which I find reprehensible and they're expensive
• 1:30:36
  they take 11% of you wait who's funding this the people who are funding the spread of Nazi content
• 1:30:41
  are the subscribers to Nazi publications sure yeah so if you are a subscriber to a paid
• 1:30:48
  sub-stack you're the money that you are sending to the author some of which goes to sub-stack
• 1:30:54
  is is uh platforming Nazi content well yeah none of us going to the Nazis none of the money
• 1:31:00
  that people send me is going to Nazis but you're giving them a pleasure yeah my commission goes to
• 1:31:04
  a company that actually is a pleasure yeah but yeah but in the way that I articulated the
• 1:31:11
  implausibility of Facebook's claim that it's not a publisher because it uses algorithms I think
• 1:31:18
  something like sub-stack is closer to um not like Facebook for me is a kind of a modern version of
• 1:31:25
  some kind of a magazine where it's actually pushing making decisions that push things at you
• 1:31:31
  with the exception of the fact that sub-stack Phil will fill in the blank space if a publisher
• 1:31:37
  doesn't want to recommend other publications to their followers such that but those spaces get filled
• 1:31:44
  with if you like this then you might also like this uh the absence of algorithmic curation
• 1:31:50
  makes me feel like sub-stack has a decent claim to take a classically liberal
• 1:31:57
  or I think he's exactly the opposite sub-stack paid righteous to come across at the beginning of
• 1:32:02
  the Nazis that would be a problem they paid they paid people 200,000 a year to come across gigantic
• 1:32:09
  I mean yes if you're on all ends of the political spectrum if you follow trans writers who have come
• 1:32:14
  off sub-stack a lot of that was because of money that went directly into the hands of anti-trans writers
• 1:32:19
  well anti-trans writers is a tricky term you they're probably talking about people like Barry Weiss
• 1:32:23
  and Andrew Sullivan and people who were driven out of the mainstream media for having slightly
• 1:32:27
  heterodox for you know for daring to question one at the time was a very a very rigid I mean I
• 1:32:33
  know people Sasha I think you know I can give you 20 people I'm just saying if you are paying
• 1:32:40
  journalists to come across to sub-stack or other writers then you are a publisher you are not
• 1:32:45
  well you're a publisher of them of those journalists but not the ones that you don't pay but
• 1:32:51
  I don't know what's your distinction um you you said that you think that sub-stack is a pure
• 1:32:56
  platform or as close to a pure platform yeah so like there were there were people who lost you know
• 1:33:01
  I was on a panel recently with Julie Sego who was one of the most um you know respected and
• 1:33:07
  longest serving opinion writers and columnists at the age the Melbourne's biggest broad sheet
• 1:33:12
  newspaper who during the social justice panic of 2022 I guess um was driven out by her peers
• 1:33:20
  for wanting to report on the controversy inside the psychiatric um community about pediatric
• 1:33:26
  transgender care right this has now now you can talk about this because it's a legitimate medical
• 1:33:33
  issue the question of whether or not a psychiatrist or a psychologist should have the right to
• 1:33:39
  look at a 12 year old who comes to them with gender dysphoria as a holistic person and consider
• 1:33:44
  also whether or not this is caused by autism or whether there are social contagion effects or
• 1:33:48
  something like that at the time this discussion that was taking place within the medical community
• 1:33:52
  itself about what do you do in these scenarios and do you fast track them directly into um trans care
• 1:33:59
  or do you look at other things that controversy she wanted to write about and that was enough to have
• 1:34:04
  her um driven out of the publication there was a what do you mean by driven out was she fired yeah
• 1:34:10
  on what ground uh she I mean her position there was untannable because she was unable to she
• 1:34:19
  she refused to not have her piece published basically so did she resign yeah uh like if you
• 1:34:27
  want to get into the minutiae of every case that I bring up then we can go down those rabbit holes
• 1:34:31
  you're saying she's driven out I want to know if she was fired uh I don't know
• 1:34:36
  um so it was if you're suggesting that there was not a massively sensorious environment in which
• 1:34:42
  people's lives were made untannable I can tell you my story at the public broadcaster I can tell
• 1:34:47
  you Barry Wyser's story at the New York Times great Linda and often tells me about history
• 1:34:51
  of ruining his life in his marriage well you know there are obviously degrees sure there are
• 1:34:56
  degrees of controversy about a sub stack as a platform why why do you think that sub stack
• 1:35:03
  specifically as broadcast like an email newsletter sending platform is preferable over other
• 1:35:11
  similar products like button down on ghost why why I've never heard a button down I've never heard
• 1:35:16
  a ghost so why so yeah so why do you why do you favor sub stack I guess I'm asking um because they
• 1:35:24
  I think they have a I think they're just more future looking so I was gonna say they have a
• 1:35:31
  commitment to um audio and video but that's newer so we just use it as a publication tool to
• 1:35:38
  deliver audio content and create community and it's possible to create um there's a there's a
• 1:35:45
  network effect of like-minded intellectually curious people who will discover us through uh not
• 1:35:54
  necessarily overt recommendations but cross collaborations with other publishers so it has the
• 1:36:00
  sense it has a it has an early internet feel of like excitement and buzz not the company itself
• 1:36:07
  but the people who are publishing on the platform who are in my universe and that's necessarily a
• 1:36:13
  occurred I mean that is a garden in a sense it's like a series of interlocking um play spaces or
• 1:36:20
  gardens where people who are interested in similar things are able to communicate without the
• 1:36:26
  uh the editorial uh restrictions of legacy media and that's a bad thing into good thing right
• 1:36:32
  so a lot of a lot of things that I could say on this podcast I wouldn't have been able to say
• 1:36:37
  on the public broadcaster in my case because I think I have a judicious journalistic temperament
• 1:36:42
  that's probably a good thing because I'm saying things that are provocative or controversial without
• 1:36:47
  veering into misinformation or hate speech in other circumstances the lack of guardrails and the
• 1:36:53
  lack of an editor and the you know the lack of a board can go horribly wrong and so you'll end
• 1:37:00
  up with people who are saying things that are either uh mildly provocative in a useful way like I
• 1:37:07
  would argue Julie say go or any other quote unquote anti-trans person was at that time they're not
• 1:37:12
  anti-trans um or it could be someone who's literally peddling in bigotry and hate speech or it could
• 1:37:19
  be any own artsy um I'm a bit ambivalent about what like it'd be a case by case thing I mean
• 1:37:28
  I would be perfectly comfortable for sub stack to have a rule where they say you're not allowed
• 1:37:33
  to promote narcissism on sub stack like that would be as the ground as the ground son of Holocaust
• 1:37:38
  survivors myself seems reasonable to me um the the devil would be in the details and the question
• 1:37:46
  I mean I think I think my concerns with sub stack are I disagree with their stated politics which is
• 1:37:52
  they have had offensive content brought to their attention several times and they have deliberately
• 1:37:57
  doubled down and said we are platforms don't publishes we're not interested in but to be honest
• 1:38:03
  social yeah given what I've lived through in how group thinky and honestly sensorious the legacy
• 1:38:12
  media has been to the past but my that relieves me because it means that I can I can explore any
• 1:38:17
  intellectual you know area that I want to and I'm not going to get a tap on the shop and I think
• 1:38:22
  the other my other consumer sub stackers they are on their way to becoming another world garden
• 1:38:28
  and so there's a bunch of as I say sort of dark patterns associated with the growth
• 1:38:33
  hacking engine with bringing people onto the app with driving to a social media sort of
• 1:38:39
  framework like when you go into the app now it's sort of all about your latest comments and so on
• 1:38:44
  so they're building a social infrastructure around the newsletters they make it very very difficult
• 1:38:49
  for you to export your followers because the audience is there's not yours and those are all
• 1:38:55
  patterns that we've seen in the past and I just don't I don't trust that it'll worry me if they
• 1:39:00
  algorithm if they like render the notes and comments and all that sort of stuff uh to
• 1:39:07
  become an algorithmic feed yeah um that'll be a tipping point for me um or at least overtly
• 1:39:13
  algorithmic you know they might be downloading things that are completely bad shit crazy at the
• 1:39:16
  moment but um which I wouldn't say um you've got a uh a nice analogy that I've heard about the
• 1:39:23
  1300-year-old onset in uh in Japan in contrast to the ethos of modern tech yeah I wish I could
• 1:39:31
  credit that it was mine but it was actually my friend Natasha Lampard who uh gave a talk at
• 1:39:36
  webstalk about this a few years ago but yeah she was talking about this onset which is um the
• 1:39:40
  world's oldest company you know it's been in existence for 400 generations and and she was using
• 1:39:47
  it as an example of um a contrast with Silicon Valley's mindset on exponential growth in exit
• 1:39:54
  you know like it's always what people talk about and start a plan is what's your exit strategy
• 1:39:57
  or someone's had a billion dollar exit and she was she was saying you know what if we
• 1:40:02
  what if we thought about an exist strategy rather than an exit strategy and what would it look
• 1:40:06
  like to build companies that lasted instead of being built to be sold um and I think there's
• 1:40:13
  something so interesting in that idea that the Silicon Valley starts is always about uh
• 1:40:21
  accelerating things as much as you can and that's what drives a lot of these start patents so
• 1:40:26
  you know venture capital investment needs to see a 10x return over you know a 10 year period
• 1:40:32
  otherwise they're not investing and they're going to hold your feet to the fire about that growth
• 1:40:38
  over that period of time how do you get growth well if you're a social media platform you need
• 1:40:43
  engaged users how do you get engagement through sticky algorithmic behavior that encourages people
• 1:40:49
  to be shown information that they are raged up about like it's just a it's a vicious cycle
• 1:40:56
  and so I like to think that if if we could change our thinking about how we build
• 1:41:03
  companies platforms tools everything like what we're making instead of thinking how do I build a
• 1:41:08
  unicorn which we know has this kind of awful effect that comes with it and instead like how do I
• 1:41:15
  make something that's useful for me for my friends for my family for my neighborhood for my
• 1:41:20
  community um and it just works at that scale you know like it doesn't it doesn't have to be
• 1:41:28
  massive it can just be healthy and last question then how do we make it healthy in our own
• 1:41:35
  lives like what would you say to people who for example have young children who have inevitably
• 1:41:40
  going to get obsessed with tech talk and whether they want to or not are going to their world view
• 1:41:45
  is going to be shaped and their understanding of news events is going to be shaped by influencers
• 1:41:49
  yeah I um I mean it's a great question I've currently got an open note see note taking
• 1:41:55
  which in which I'm just pasting every link I see to an article which says either phones are terrible
• 1:42:02
  for teens or phones are fine and honestly every link goes flip-flop like I read one study it's like
• 1:42:09
  phones are destroying the nation's health and then the next one is like actually study shows that
• 1:42:14
  teenage anxiety is not connected with phones and you know it just goes back and forth back and forth
• 1:42:19
  um so I don't I don't have an answer to that what I do think is interesting is and this is the
• 1:42:23
  challenge I mean just jump in sorry I don't think the problem is fine it's right I mean the problem
• 1:42:27
  is the algorithm right there but I would I cannot believe that it is not well I see what my son is
• 1:42:34
  like when he's playing an addictive video game versus when he's doing something constructive
• 1:42:39
  you know or interacting and this is a discussion after Netflix dropped our lessons right and a lot
• 1:42:44
  of people watch that many series and where well it's the phones and it's like missing the point
• 1:42:49
  of the show entirely yeah which is that it's not the phones it's the entire world that they're
• 1:42:53
  steeped in yes so yeah but I do one of the things that I did when I was preparing this talk this
• 1:43:01
  for this week was I was thinking about how we used to build for the web and my nephew Josh is
• 1:43:07
  15 he's a really keen football player and he wanted to start coaching other kids and so we
• 1:43:11
  needed a website and he'd ask me what service he should sign up for to make his website and I
• 1:43:16
  suddenly thought because I was immersed in this nostalgia trip I was like why don't we make it
• 1:43:20
  like let's make it yourself and so we just sat down I used to chat GPT to start because I knew
• 1:43:29
  if I tried to explain HTML to him his eyes would glaze over so I said just ask the LLM for some
• 1:43:34
  basic coach for a football coaching website and I showed him how to save it done locally and open
• 1:43:39
  it on his machine and you know his jaw dropped when the index file opened up and he was amazed when
• 1:43:45
  seconds later it had some CSS within 20 minutes he was debugging the mistakes that the era was making
• 1:43:50
  and it was making mistakes and within an hour he had a site that had started to look and feel like
• 1:43:56
  him and he had made it and there was something so energising and creative about that and I think
• 1:44:03
  we've just forgotten that the good internet is still out there it's on the other side of the walls
• 1:44:09
  or just turn the computer off and go outside yeah kick the ball park
• 1:44:13
  both things can be true thank you for talking to us and Greg thank you it's been great