Latest / Elon Musk Podcast / Elon Musk Talks about the future of social media and AI

Transcript

Elon Musk Talks about the future of social media and AI · Elon Musk Podcast

• 0:01
  Hey everybody, welcome back to the Elon Musk Podcast.
• 0:05
  This is a show where we discuss the critical Crossroads, The
• 0:08
  Shape, SpaceX, Tesla X, The Boring Company, and Neuralink,
• 0:13
  and I'm your host, Will Walden. If you want uninterrupted
• 0:17
  episodes of the Elon Musk podcast, please go to
• 0:20
  clubelon.supercast.com to find out how there's a link in the
• 0:26
  show notes. I'm saying this is something we
• 0:29
  should be quite concerned about, because if Robot can follow you
• 0:32
  anywhere, then you know what? If they just one day get a
• 0:37
  software update and they're not so friendly anymore, then we've
• 0:41
  got a James Cameron movie on our heads, right?
• 0:45
  It's not the. It's not the obvious place you'd
• 0:47
  go to, but. Maybe that could be one of the
• 0:48
  tests for the AI. We should say like Blank is your
• 0:52
  favorite James Cameron movie. Fill in the blank.
• 1:00
  Well, I think we shouldn't trust it that much, actually.
• 1:06
  It is actually quite, quite a significant challenge because
• 1:08
  we're getting to the point where even open source AI can pass
• 1:11
  human capture tests. So, you know, this is, are you a
• 1:14
  human? Identify all the traffic lights
• 1:16
  in this picture. You're like, OK, yeah, it's
• 1:19
  going to have a no problem doing that.
• 1:21
  In fact, it'll do it better than human and faster than human.
• 1:24
  So we're like how do you, you know, at the point which is
• 1:27
  better, a better human, better passing human tests than humans,
• 1:31
  then well, what tests actually make sense?
• 1:34
  That is a real problem and I don't actually have a good
• 1:36
  solution to it. That one of the things we're
• 1:38
  trying to figure out on the X platform is how to deal with
• 1:42
  that because it really we really are at the point where even with
• 1:47
  with open source you know readily available AI you don't
• 1:50
  need to be sort of leading in the field.
• 1:53
  You can actually be better than humans at passing these tests
• 1:56
  And that's sort of why we think well perhaps we should sort of
• 1:59
  charge up a dollar or a pound a year.
• 2:01
  It's a very tiny amount of money but it's it's still it still
• 2:04
  makes it privatively expensive to make a million bahts.
• 2:08
  So and especially if you need a million payment methods then you
• 2:12
  run out of sort of stolen credit cards pretty quickly.
• 2:18
  So that that's that's sort of where we're thinking like we
• 2:21
  might have to sort of just charge some very tiny amount of
• 2:23
  money .3 cents a day effectively to deal with the the onslaught
• 2:30
  of AI powered bonds. And if and that that is, that is
• 2:35
  still a growing problem but it will be I think perhaps an
• 2:38
  insurmountable problem next year.
• 2:40
  So and then you have to worry about, well manipulation of
• 2:45
  information is making something seem very popular when in fact
• 2:48
  it is not because it's getting boosted by all these likes and
• 2:52
  repos from AI powered bots. So that's why I sort of think so
• 2:57
  somewhat inevitably it leads to some small payment in order to
• 3:02
  dramatically increase the cost of a bot.
• 3:06
  So I think, I think frankly I think probably any social media
• 3:10
  system that doesn't do that will simply be overrun by bots.
• 3:13
  Exactly to your point. I mean you the the images, it's
• 3:15
  it's you don't even need to steal somebody's picture because
• 3:18
  that's traceable. But you can actually just say
• 3:21
  create a new a new image of a person realistic looking that
• 3:24
  doesn't exist and and then create a a biography realistic
• 3:28
  that doesn't exist and do that on mass and and practically the
• 3:32
  only way to be able to tell us that the grammar's too good.
• 3:36
  Dick giveaway no typos. Come on.
• 3:38
  Yeah. I mean some of it is is quite
• 3:40
  entertaining like the Pope in the puffa jacket.
• 3:42
  Have you seen that one That's amazing.
• 3:45
  But I mean I still run into people who who who think that's
• 3:47
  real. I'm like well what are the odds
• 3:50
  he's wearing a puffa jacket in July in Rome you know be
• 3:54
  sweating. But it actually looked quite
• 3:56
  quite dashing of the in fact I think AI fashion is going to be
• 4:00
  a real thing. So I don't remember doom and
• 4:02
  gloom like we learned in the most interesting times and I
• 4:04
  think this is it is you know 8 like 80% likely to be good and
• 4:09
  20% bad. I think we're, if we're
• 4:11
  cognizant and and careful about the bad part on balance actually
• 4:14
  it will be the future that we want or the for the future that
• 4:18
  is preferable. And and it actually will be a
• 4:22
  somewhat of a level an equalizer in the sense that you know, I
• 4:26
  think everyone will have access to goods and services and
• 4:28
  education. And so, you know, I think
• 4:31
  probably it leads to more human happiness.
• 4:34
  So I I guess I'd probably leave on, on an optimistic note, yeah.
• 4:39
  That's it. Well, that's it.
• 4:40
  That is a great note to end on. Hey, thank you so much for
• 4:43
  listening today. I really do appreciate your
• 4:45
  support. If you could take a second and
• 4:46
  hit the subscribe or the follow button on whatever podcast
• 4:49
  platform that you're listening on right now, I greatly
• 4:52
  appreciate it. It helps out the show
• 4:54
  tremendously and you'll never miss an episode.
• 4:57
  And each episode is about 10 minutes or less to get you
• 5:00
  caught up quickly. Please, if you want to support
• 5:03
  the show even more, go to Atreoncom, Stage Zero, and
• 5:09
  please take care of yourselves and each other and I'll see you
• 5:11
  tomorrow.