WEBVTT - The BS Bubble

0:00:02.800 --> 0:00:03.560
<v Speaker 1>Zone Media.

0:00:05.480 --> 0:00:18.079
<v Speaker 2>Hi, I'm at Zitron and you're listening to Better Offline.

0:00:20.079 --> 0:00:23.400
<v Speaker 2>As a reminder, you can now buy glorious Better Offline merchandise.

0:00:23.400 --> 0:00:25.159
<v Speaker 2>There's a link to it in the episode notes. The

0:00:25.200 --> 0:00:28.120
<v Speaker 2>t shirts, the tumblers, the toe bags, all that shit.

0:00:28.200 --> 0:00:31.880
<v Speaker 2>It's lovely. You're gonna love it, buy it today. But

0:00:32.000 --> 0:00:34.360
<v Speaker 2>if I'm honest today, I'm actually a little bit pissed off.

0:00:34.360 --> 0:00:36.239
<v Speaker 2>And that's why we've got a two part episode this

0:00:36.320 --> 0:00:38.880
<v Speaker 2>week about how fucking stupid the AI boom has become.

0:00:39.000 --> 0:00:42.200
<v Speaker 2>I wrote fascicle in the script, and I'm gonna be honest.

0:00:42.240 --> 0:00:44.319
<v Speaker 2>They need to be a little more pointed because I've

0:00:44.320 --> 0:00:46.760
<v Speaker 2>written tens of thousands of words about this. Now, I

0:00:46.880 --> 0:00:49.880
<v Speaker 2>recorded hours upon hours of podcast and still to this day,

0:00:50.240 --> 0:00:52.440
<v Speaker 2>people are babbling about the AI revolution. Is the sky

0:00:52.520 --> 0:00:54.680
<v Speaker 2>rains blood and crevices open in the fucking earth, dragging

0:00:54.680 --> 0:00:58.080
<v Speaker 2>houses and cars and dovestigated animals into them ores. Things

0:00:58.120 --> 0:01:01.960
<v Speaker 2>are astronomically fucked outside, Yet the tech media continues to

0:01:01.960 --> 0:01:03.680
<v Speaker 2>tell me to get my swimming trunks on and take

0:01:03.680 --> 0:01:06.560
<v Speaker 2>a long, nice dip in the fucking pool. As you

0:01:06.600 --> 0:01:08.120
<v Speaker 2>can tell, this is going to be a little less

0:01:08.120 --> 0:01:11.280
<v Speaker 2>reserved than usual. I've just I'm a little bit frustrated.

0:01:12.200 --> 0:01:14.319
<v Speaker 2>I don't know why I'm the one saying this, and

0:01:14.360 --> 0:01:17.080
<v Speaker 2>I frequently feel with the I a part time blogger

0:01:17.120 --> 0:01:21.119
<v Speaker 2>and podcaster. I'm writing the things that I'm writing. Since

0:01:21.160 --> 0:01:23.400
<v Speaker 2>I've put out the newsletter open ai as the systemic

0:01:23.520 --> 0:01:25.360
<v Speaker 2>risk of the tech industry, and actually it's a couple

0:01:25.360 --> 0:01:27.360
<v Speaker 2>of weeks back, I did the two parter about it too.

0:01:27.560 --> 0:01:29.760
<v Speaker 2>I've heard nothing in response, As was the case with

0:01:29.800 --> 0:01:32.040
<v Speaker 2>how does open ai survive and how open ai is

0:01:32.120 --> 0:01:36.480
<v Speaker 2>bad business. There just seems to be little concern or

0:01:36.520 --> 0:01:38.840
<v Speaker 2>belief that there's any kind of risk at the heart

0:01:38.920 --> 0:01:41.200
<v Speaker 2>of AI and open ai in particular. And there are

0:01:41.200 --> 0:01:44.200
<v Speaker 2>companies that spent nine billion dollars in twenty twenty four

0:01:44.280 --> 0:01:47.280
<v Speaker 2>to lose five billion dollars. Well, I'd love to add

0:01:47.280 --> 0:01:50.080
<v Speaker 2>a because here, if not, because it's important to be

0:01:50.080 --> 0:01:52.800
<v Speaker 2>intellectually honest and represent views that directly can trast my own,

0:01:52.840 --> 0:01:56.000
<v Speaker 2>even if I do in somewhat sarcastic and sardonic fashion.

0:01:56.200 --> 0:01:58.680
<v Speaker 2>Nobody seems to actually have a cogent response to how

0:01:58.680 --> 0:02:01.520
<v Speaker 2>they write this ship hard fork are case in Newton

0:02:01.680 --> 0:02:04.440
<v Speaker 2>throwing a full scale Psycho Tante on a podcast and

0:02:04.480 --> 0:02:08.240
<v Speaker 2>saying I'm wrong because inference costs are coming down. Inference,

0:02:08.320 --> 0:02:10.240
<v Speaker 2>by the way, is when an AI takes an input

0:02:10.320 --> 0:02:13.400
<v Speaker 2>and produces an output. It's the calculations that take place

0:02:13.520 --> 0:02:16.880
<v Speaker 2>right before Google's generativai assistant attributes the Voltaire quote to

0:02:17.120 --> 0:02:20.400
<v Speaker 2>Michael Jackson or says that black tar heroin, when enjoyed

0:02:20.400 --> 0:02:23.440
<v Speaker 2>in moderation, can help you lose weight. Newton is a

0:02:23.560 --> 0:02:27.000
<v Speaker 2>nakedly captured booster that ran an infographic from Anthropic a

0:02:27.000 --> 0:02:29.040
<v Speaker 2>few weeks ago, likes of which I haven't seen since

0:02:29.040 --> 0:02:31.200
<v Speaker 2>twenty thirteen. It was telling you all the ways that

0:02:31.240 --> 0:02:33.720
<v Speaker 2>people use Genera ivai. It looks like some shit from

0:02:33.840 --> 0:02:37.000
<v Speaker 2>I don't know, early day Mashable. No offense Christina, and

0:02:37.040 --> 0:02:40.560
<v Speaker 2>they essentially treat this company propaganda as gospel. But he's

0:02:40.600 --> 0:02:43.280
<v Speaker 2>really far from the only one with a flimsy attachment

0:02:43.320 --> 0:02:46.960
<v Speaker 2>to reality. The information a publication that genuinely does some

0:02:47.040 --> 0:02:49.720
<v Speaker 2>great stuff, which makes it even more heartbreaking to say this.

0:02:49.840 --> 0:02:52.520
<v Speaker 2>Ran a piece in early April that made me even

0:02:52.520 --> 0:02:55.760
<v Speaker 2>more furious than usual, claiming that open ai was forecasting

0:02:55.800 --> 0:02:57.840
<v Speaker 2>revenue topping one hundred and twenty five billion dollars in

0:02:57.880 --> 0:03:01.840
<v Speaker 2>twenty twenty nine, based on selling agents and monetizing free

0:03:01.919 --> 0:03:06.000
<v Speaker 2>users as a driver to higher revenue agents. I should

0:03:06.000 --> 0:03:08.720
<v Speaker 2>add AI systems that can interact with other systems and

0:03:08.760 --> 0:03:11.959
<v Speaker 2>do stuff. So an AI that can order pizza from

0:03:12.000 --> 0:03:14.240
<v Speaker 2>door dash for you is that's an example of an agent.

0:03:14.280 --> 0:03:15.840
<v Speaker 2>And when I say it can order piece of you,

0:03:16.280 --> 0:03:18.960
<v Speaker 2>I am talking entirely theoretically. Is they cannot do this

0:03:19.040 --> 0:03:22.239
<v Speaker 2>right now and may never be able to do so. Indeed,

0:03:22.360 --> 0:03:25.480
<v Speaker 2>the whole agent thing is just what we wish AI

0:03:25.760 --> 0:03:28.680
<v Speaker 2>was and it actually doesn't work. And the piece reported

0:03:28.720 --> 0:03:31.760
<v Speaker 2>based on things and I quote told to some potential

0:03:31.800 --> 0:03:35.120
<v Speaker 2>and current investors, takes great pains to accept literally everything

0:03:35.120 --> 0:03:38.200
<v Speaker 2>that open ai says is perfectly reasonable, if not gospel,

0:03:38.240 --> 0:03:42.240
<v Speaker 2>even if said things make absolutely no goddamn sense. So,

0:03:42.600 --> 0:03:46.080
<v Speaker 2>according to the information's reporting, open ai expects agents and

0:03:46.200 --> 0:03:49.240
<v Speaker 2>new products, and both of those are quotes to contribute

0:03:49.280 --> 0:03:52.360
<v Speaker 2>tens of billions of dollars of revenue, both in the

0:03:52.400 --> 0:03:56.080
<v Speaker 2>near term somehow contributing three billion dollars in revenue this year,

0:03:56.120 --> 0:03:57.760
<v Speaker 2>which I will get to in a little bit, and

0:03:57.800 --> 0:04:00.720
<v Speaker 2>in the long term with an egregious twenty five billion

0:04:00.760 --> 0:04:04.360
<v Speaker 2>dollars in revenue in twenty twenty nine, even projected to

0:04:04.400 --> 0:04:08.360
<v Speaker 2>come from just new products. If you're wondering what those

0:04:08.400 --> 0:04:11.160
<v Speaker 2>new products might be, I am two because the information

0:04:11.240 --> 0:04:13.360
<v Speaker 2>doesn't seem to know and instead of saying open ai

0:04:13.520 --> 0:04:15.400
<v Speaker 2>has no idea what the fuck they're talking about and

0:04:15.480 --> 0:04:19.120
<v Speaker 2>is just saying stuff, the outlet continues to publish things

0:04:19.120 --> 0:04:22.080
<v Speaker 2>with the kind of empty optimism that's indistinguishable from GPT

0:04:22.520 --> 0:04:26.200
<v Speaker 2>generated LinkedIn posts. Must be clear, the information isn't generating

0:04:26.200 --> 0:04:30.040
<v Speaker 2>their articles they're writing and fresh. I want to be really,

0:04:30.040 --> 0:04:33.599
<v Speaker 2>really clear about something we aren't in nearly in May

0:04:33.320 --> 0:04:35.520
<v Speaker 2>twenty five, and indeed one of these will come out

0:04:35.720 --> 0:04:39.320
<v Speaker 2>actually in May. The second part, I see no evidence

0:04:39.320 --> 0:04:41.920
<v Speaker 2>that open ai even has a marketable agent product they

0:04:41.920 --> 0:04:44.680
<v Speaker 2>can sell, let alone one will it will make three

0:04:44.720 --> 0:04:49.320
<v Speaker 2>billion goddamn dollars off of and they definitely are not

0:04:49.400 --> 0:04:51.640
<v Speaker 2>going to do so in the next six or seven months.

0:04:51.760 --> 0:04:55.240
<v Speaker 2>Oh my, for context, let's triple the revenue of open

0:04:55.279 --> 0:04:58.520
<v Speaker 2>ai that they made reportedly at least from selling access

0:04:58.560 --> 0:05:01.520
<v Speaker 2>to their models via at APIs essentially allowing third party

0:05:01.520 --> 0:05:04.160
<v Speaker 2>companies to use GPT in their apps in the entirety

0:05:04.200 --> 0:05:07.160
<v Speaker 2>of twenty twenty four, and those APIs and models actually

0:05:07.160 --> 0:05:09.960
<v Speaker 2>exist in a meaningful sense, as opposed to whatever the

0:05:09.960 --> 0:05:13.440
<v Speaker 2>far copenai is half baked last agents stuff is. In fact, no, no, no,

0:05:13.440 --> 0:05:16.320
<v Speaker 2>no no. I'm not going to be mean, not would

0:05:16.320 --> 0:05:19.320
<v Speaker 2>be calm, be normal. I'm going to explain exactly what

0:05:19.360 --> 0:05:22.200
<v Speaker 2>the information is reporting in an objective way, because writing

0:05:22.240 --> 0:05:24.600
<v Speaker 2>it out really shows how silly it all sounds. I

0:05:24.640 --> 0:05:26.800
<v Speaker 2>am going to rate they believe a lot, because I

0:05:26.960 --> 0:05:30.520
<v Speaker 2>must be clear how stupid this is. Now, according to

0:05:30.560 --> 0:05:32.840
<v Speaker 2>the Information's reporting, they believe the open AI will make

0:05:32.880 --> 0:05:35.240
<v Speaker 2>three billion dollars in twenty twenty five from selling access

0:05:35.240 --> 0:05:37.800
<v Speaker 2>to its agents. This appears to come from soft Bank,

0:05:37.839 --> 0:05:39.960
<v Speaker 2>which has said it will buy three billion dollars worth

0:05:40.000 --> 0:05:42.880
<v Speaker 2>of open ai products annually. Earlier this year, we got

0:05:42.920 --> 0:05:45.600
<v Speaker 2>a bit of extra information about how soft Bank will

0:05:45.680 --> 0:05:47.960
<v Speaker 2>use these products. It plans to create a system called

0:05:48.000 --> 0:05:52.599
<v Speaker 2>Crystal Intelligence that CRISTL and it's one of the most

0:05:52.600 --> 0:05:54.400
<v Speaker 2>generic names I've ever seen, and it will be a

0:05:54.480 --> 0:05:57.520
<v Speaker 2>kind of general purpose AI agent platform for big enterprises.

0:05:58.000 --> 0:06:00.960
<v Speaker 2>The exact specifics of that will shock you, and that

0:06:01.000 --> 0:06:03.560
<v Speaker 2>there are none, but SoftBank intends to use the technology

0:06:03.800 --> 0:06:08.040
<v Speaker 2>internally across its various portfolio companies, as well as market

0:06:08.560 --> 0:06:11.680
<v Speaker 2>it to other large enterprise companies in Japan. I still

0:06:11.720 --> 0:06:14.279
<v Speaker 2>do not know what the fuck this is Crystal Intelligence

0:06:14.600 --> 0:06:17.880
<v Speaker 2>billions of dollars, billions of dollars, and they just don't.

0:06:17.920 --> 0:06:20.400
<v Speaker 2>They can't even describe what it is, just saying yeah,

0:06:20.400 --> 0:06:25.640
<v Speaker 2>it'll be an agent platform that does stuff with your business,

0:06:25.800 --> 0:06:28.600
<v Speaker 2>Like does that sound good? Can I have? I need

0:06:28.640 --> 0:06:32.880
<v Speaker 2>forty billion dollars? I need forty billion dollars, give me okay.

0:06:33.960 --> 0:06:35.880
<v Speaker 2>I also want to add that the Information can't seem

0:06:35.880 --> 0:06:38.839
<v Speaker 2>to keep its story straight on this issue. Back in February,

0:06:38.839 --> 0:06:41.120
<v Speaker 2>they reported that open AI would make three billion dollars

0:06:41.120 --> 0:06:44.200
<v Speaker 2>in revenue only from agents, with a big beautiful chart

0:06:44.240 --> 0:06:47.040
<v Speaker 2>that said three billion dollars would come from it, only

0:06:47.080 --> 0:06:49.120
<v Speaker 2>to add that it would be soft Bank using open

0:06:49.160 --> 0:06:52.839
<v Speaker 2>AIS products across its companies. Based on these numbers, it

0:06:52.880 --> 0:06:55.200
<v Speaker 2>seems like SoftBank will be the only customer for open

0:06:55.240 --> 0:06:58.000
<v Speaker 2>AIS agents. Well, this most likely won't be the case,

0:06:58.000 --> 0:07:00.000
<v Speaker 2>and it isn't because it excludes anyone willing to pay

0:07:00.000 --> 0:07:02.200
<v Speaker 2>a few bucks to test it out. It nonetheless doesn't

0:07:02.200 --> 0:07:04.640
<v Speaker 2>signal good things for agents as a mass market product,

0:07:05.320 --> 0:07:08.120
<v Speaker 2>not that there were any good signals beforehand. Though agents

0:07:08.120 --> 0:07:09.920
<v Speaker 2>do not exist as a product that can be sold

0:07:09.960 --> 0:07:13.040
<v Speaker 2>at scale. Yes, open ai teasted Operator it's first agent

0:07:13.080 --> 0:07:15.080
<v Speaker 2>at the start of the year, but it doesn't seem

0:07:15.120 --> 0:07:18.920
<v Speaker 2>to be able to do anything. The Information's own reporting

0:07:18.920 --> 0:07:22.680
<v Speaker 2>from mid April highlighted how OpenAIS operate a agent struggled

0:07:22.720 --> 0:07:25.320
<v Speaker 2>with comparison shopping on financial products and that's a quote.

0:07:25.680 --> 0:07:28.000
<v Speaker 2>And how Operator or other agents are and I quote

0:07:28.000 --> 0:07:30.720
<v Speaker 2>again tripped by pop ups or logins as well as

0:07:30.840 --> 0:07:33.800
<v Speaker 2>prompts asking for email addresses and phone numbers for marketing

0:07:33.960 --> 0:07:40.400
<v Speaker 2>purpose purposes, which I think accurately describes most websites. And

0:07:40.560 --> 0:07:42.760
<v Speaker 2>just to summarize from everything I've said, the Information is

0:07:42.800 --> 0:07:45.080
<v Speaker 2>saying that the above product will make open ai three

0:07:45.160 --> 0:07:48.160
<v Speaker 2>billion dollars by the end of twenty twenty five. Sounds

0:07:48.280 --> 0:07:52.760
<v Speaker 2>very real to me, sounds extremely real. I love that

0:07:52.800 --> 0:07:56.680
<v Speaker 2>the business media just prints this. I love this. I

0:07:56.720 --> 0:08:01.520
<v Speaker 2>love this so much. I'm having so much fun. Jesus Christ.

0:08:01.600 --> 0:08:04.000
<v Speaker 2>According to The Information's reporting, they believe that open ai

0:08:04.080 --> 0:08:06.640
<v Speaker 2>will basically double revenue every single year for the next

0:08:06.720 --> 0:08:09.240
<v Speaker 2>four years and make thirteen billion dollars in revenue twenty

0:08:09.280 --> 0:08:11.640
<v Speaker 2>twenty five, more than doubling that to twenty nine billion

0:08:11.680 --> 0:08:14.480
<v Speaker 2>dollars in twenty twenty six, nearly doubling that to fifty

0:08:14.480 --> 0:08:17.040
<v Speaker 2>four billion dollars in twenty twenty seven, and nearly doubling

0:08:17.120 --> 0:08:19.840
<v Speaker 2>that again to eighty six billion dollars in twenty twenty eight,

0:08:19.840 --> 0:08:22.840
<v Speaker 2>and eventually leveling out a ridiculous one hundred and twenty

0:08:22.840 --> 0:08:25.680
<v Speaker 2>five billion dollars of revenue in twenty twenty nine. Said

0:08:25.720 --> 0:08:28.520
<v Speaker 2>revenue estimates as of twenty twenty six includes billions of

0:08:28.640 --> 0:08:32.680
<v Speaker 2>dollars of new products that include free monetization free user

0:08:32.720 --> 0:08:35.560
<v Speaker 2>monetization either and if you're wondering what that means, I

0:08:36.080 --> 0:08:40.079
<v Speaker 2>also am. The information does not explain JESSICLLESSI must have

0:08:40.120 --> 0:08:42.640
<v Speaker 2>been busy being horrible to people that work for her.

0:08:42.960 --> 0:08:45.080
<v Speaker 2>They do, however, say that open ai will start, and

0:08:45.080 --> 0:08:48.160
<v Speaker 2>I'm quoting this won't start generating much revenue from free

0:08:48.240 --> 0:08:51.000
<v Speaker 2>users and other products until next year. That's twenty twenty six.

0:08:51.400 --> 0:08:54.559
<v Speaker 2>In twenty nine, and I'm still quoting. However, it projects

0:08:54.600 --> 0:08:57.240
<v Speaker 2>revenue from free users and other products will reach twenty

0:08:57.240 --> 0:09:00.120
<v Speaker 2>five billion dollars, a one fifth of all revenue, and

0:09:00.120 --> 0:09:05.679
<v Speaker 2>then adds that shopping is another potential avenue. You still

0:09:05.720 --> 0:09:08.600
<v Speaker 2>probably don't know what they're doing, and neither do iron

0:09:08.960 --> 0:09:13.040
<v Speaker 2>I have driven myself insane reading about this. I really

0:09:13.080 --> 0:09:16.360
<v Speaker 2>cannot express my disgust about how willing publications are to

0:09:16.400 --> 0:09:19.800
<v Speaker 2>blindly published projections like these, especially when they're all so stupid.

0:09:20.160 --> 0:09:23.280
<v Speaker 2>Let me just read this to you, all right, and

0:09:23.360 --> 0:09:26.720
<v Speaker 2>I quote. Open ai has already begun experimenting with launching

0:09:26.720 --> 0:09:29.800
<v Speaker 2>software features for shopping. Starting in January, some users can

0:09:29.840 --> 0:09:32.920
<v Speaker 2>access web browsing agent Operator as part of their Prochat

0:09:32.960 --> 0:09:36.440
<v Speaker 2>GBT subscription to here to order groceries from Instacart and

0:09:36.480 --> 0:09:40.600
<v Speaker 2>make restaurant reservations on open table. Just want to be clear.

0:09:40.640 --> 0:09:43.600
<v Speaker 2>This is a few episodes ago I mentioned Casey Newton

0:09:43.679 --> 0:09:47.480
<v Speaker 2>not even being able to say this worked. I just

0:09:47.480 --> 0:09:49.400
<v Speaker 2>want to be really clear as well what the information

0:09:49.440 --> 0:09:52.680
<v Speaker 2>is saying. So they're saying that this experimental software launched

0:09:52.679 --> 0:09:55.199
<v Speaker 2>to an indeterminate amount of people that barely works, is

0:09:55.240 --> 0:09:58.199
<v Speaker 2>going to make open Ai three billion dollars in twenty

0:09:58.240 --> 0:10:01.000
<v Speaker 2>twenty five, and then somehow this is going to lead

0:10:01.040 --> 0:10:03.160
<v Speaker 2>to open ai making twenty nine billion dollars in twenty

0:10:03.200 --> 0:10:05.240
<v Speaker 2>twenty six, and then they're can eventually be up to

0:10:05.280 --> 0:10:08.600
<v Speaker 2>one hunch and twenty five billion dollars. What the fuck? How? How?

0:10:08.760 --> 0:10:11.360
<v Speaker 2>What fucking universe are we all living in? There's no

0:10:11.520 --> 0:10:13.480
<v Speaker 2>proof that open ai can do this other than the

0:10:13.480 --> 0:10:15.080
<v Speaker 2>fact it has a lot of users and a lot

0:10:15.120 --> 0:10:27.400
<v Speaker 2>of venture capital. In fact, I think we have real

0:10:27.480 --> 0:10:30.120
<v Speaker 2>reason to worry about whether open ai even makes its

0:10:30.160 --> 0:10:33.520
<v Speaker 2>current projections. In my last multi part episode and then

0:10:33.559 --> 0:10:35.800
<v Speaker 2>the newsletter open ai as a systemic risk for those

0:10:35.840 --> 0:10:37.480
<v Speaker 2>of you who are like to read while listening to

0:10:37.520 --> 0:10:40.599
<v Speaker 2>my fucking podcast, I wrote the Bloomberg had estimated that

0:10:40.640 --> 0:10:43.440
<v Speaker 2>open ai would triple revenue to twelve point seven billion

0:10:43.440 --> 0:10:45.400
<v Speaker 2>dollars in twenty twenty five, and based on its current

0:10:45.440 --> 0:10:48.360
<v Speaker 2>subscriber base, open ai would effectively have to double its

0:10:48.360 --> 0:10:51.840
<v Speaker 2>current subscription revenue and massively increase its API revenue to

0:10:51.920 --> 0:10:55.960
<v Speaker 2>hit these targets. These projections rely on one entity, SoftBank,

0:10:56.000 --> 0:10:59.600
<v Speaker 2>spending three billion dollars specifically on open AI's services really

0:10:59.600 --> 0:11:02.160
<v Speaker 2>shouldn't said specifically, because they keep changing what it means,

0:11:02.600 --> 0:11:05.360
<v Speaker 2>meaning that they'd have to make enough on API course,

0:11:05.400 --> 0:11:07.640
<v Speaker 2>so people plug in the models into their products to

0:11:07.679 --> 0:11:10.160
<v Speaker 2>generate more revenue. In the open Ai made in subscriptions

0:11:10.160 --> 0:11:12.960
<v Speaker 2>in the entirety of twenty twenty four and something else

0:11:13.000 --> 0:11:16.080
<v Speaker 2>that I can only describe as an act of God,

0:11:16.800 --> 0:11:18.880
<v Speaker 2>and that I admit. It seems that soft bank spending

0:11:18.880 --> 0:11:21.160
<v Speaker 2>commitment is based on usage and not like a flat

0:11:21.160 --> 0:11:23.640
<v Speaker 2>fee where Softpak just hands them three billion dollars and

0:11:23.679 --> 0:11:28.280
<v Speaker 2>gets infinite levels of access. Assuming it's the former, I'd

0:11:28.280 --> 0:11:31.120
<v Speaker 2>be stunned if soft Bank's consumption hits three billion dollars

0:11:31.160 --> 0:11:33.200
<v Speaker 2>in twenty twenty five, even with the massive cost of

0:11:33.240 --> 0:11:36.120
<v Speaker 2>the reasoning models that Crystal Intelligence will maybe be based

0:11:36.120 --> 0:11:39.559
<v Speaker 2>off of. Again, we don't know, and soft Bank announced

0:11:39.600 --> 0:11:42.920
<v Speaker 2>this deal with open Ai in February. Crystal Intelligence, if

0:11:42.920 --> 0:11:45.319
<v Speaker 2>it works, and that is possibly the most load bearing

0:11:45.480 --> 0:11:48.280
<v Speaker 2>IF of all time, will be a massive, complicated and

0:11:48.320 --> 0:11:51.680
<v Speaker 2>ambitious product. Details are vague, but from what I understand,

0:11:51.720 --> 0:11:53.640
<v Speaker 2>soft Bank wants to create an AI that handles a

0:11:53.640 --> 0:11:56.320
<v Speaker 2>bunch of varied tasks that knowledge workers do. I mean,

0:11:56.360 --> 0:11:59.880
<v Speaker 2>it's just the same marketing bullshit. It's the same thing.

0:12:00.080 --> 0:12:02.240
<v Speaker 2>It's the thing they've been lying about before. And to

0:12:02.240 --> 0:12:05.520
<v Speaker 2>be clear, open AI's agents cannot consistently do well anything

0:12:05.640 --> 0:12:08.679
<v Speaker 2>right now, what I believe is happening is that reporters

0:12:08.720 --> 0:12:11.240
<v Speaker 2>are taking open AI's rapid growth in revenue from twenty

0:12:11.240 --> 0:12:13.240
<v Speaker 2>twenty three to twenty twenty four, when they went from

0:12:13.280 --> 0:12:15.600
<v Speaker 2>like tens of millions of dollars a month in the

0:12:15.600 --> 0:12:17.960
<v Speaker 2>beginning of the twenty twenty three to three hundred million

0:12:18.000 --> 0:12:20.760
<v Speaker 2>in August twenty twenty four, genuinely a big leap. They've

0:12:20.760 --> 0:12:23.040
<v Speaker 2>taken this to mean that the company will always effectively

0:12:23.040 --> 0:12:25.920
<v Speaker 2>double or triple revenue every single year forever, with their

0:12:25.960 --> 0:12:28.720
<v Speaker 2>evidence being open ai has said that this will happen

0:12:28.720 --> 0:12:32.960
<v Speaker 2>in projections. It's bullshit. I'm sorry, it's bullshit. It's bullshit.

0:12:34.000 --> 0:12:36.880
<v Speaker 2>As I wrote before in a newsletter, it's called there's

0:12:36.920 --> 0:12:40.000
<v Speaker 2>No Ai Revolution and the accompanying episodes. At the time,

0:12:40.080 --> 0:12:43.000
<v Speaker 2>open Ai effectively is the generative AI industry, and nothing

0:12:43.040 --> 0:12:45.720
<v Speaker 2>about the rest of the generative AI industry suggests that

0:12:45.760 --> 0:12:49.520
<v Speaker 2>the revenue exists to sustain these ridiculous obscenes and frankly

0:12:49.559 --> 0:12:53.640
<v Speaker 2>fucking stupid valuations and projections. What do I mean by that?

0:12:53.679 --> 0:12:56.360
<v Speaker 2>By the way, okay, let me get into it. Chat

0:12:56.400 --> 0:12:58.840
<v Speaker 2>GPT is the only real generative AI product with any

0:12:58.920 --> 0:13:02.240
<v Speaker 2>significant usage, or rather their nearest rivals or a fraction

0:13:02.320 --> 0:13:04.480
<v Speaker 2>of said user base, or maybe I need to be

0:13:04.520 --> 0:13:07.960
<v Speaker 2>a little bit blunter. If anyone held at Google Gemini

0:13:08.120 --> 0:13:11.000
<v Speaker 2>user conference, all the attendees could probably share a cab.

0:13:11.960 --> 0:13:14.760
<v Speaker 2>Believing the open AI growth myth and yes, reporting it

0:13:14.800 --> 0:13:18.040
<v Speaker 2>objectively is both endorsing and believing these numbers is engaging

0:13:18.040 --> 0:13:20.840
<v Speaker 2>in childlike logic where you take one event, which is

0:13:20.880 --> 0:13:24.040
<v Speaker 2>open AI's revenue grew seventeen hundred percent from twenty twenty

0:13:24.040 --> 0:13:27.480
<v Speaker 2>three to twenty twenty four. Wow, to mean another will

0:13:27.480 --> 0:13:29.480
<v Speaker 2>take place, which is the open a I will continue

0:13:29.520 --> 0:13:32.920
<v Speaker 2>to double revenue literally every other year. Another insane thing

0:13:32.960 --> 0:13:36.520
<v Speaker 2>to believe, and you're consciously ignoring difficult questions such as

0:13:36.800 --> 0:13:39.439
<v Speaker 2>how will they do this? And what's the total addressable

0:13:39.480 --> 0:13:42.800
<v Speaker 2>market of large language models and their associated subscriptions exactly?

0:13:42.800 --> 0:13:45.360
<v Speaker 2>And how does this company even survive when it expects

0:13:45.400 --> 0:13:47.360
<v Speaker 2>the costs of inference the triple this year to six

0:13:47.400 --> 0:13:52.440
<v Speaker 2>billion dollars alone. Wait wait, wait, sorry, sorry, I really

0:13:52.480 --> 0:13:54.200
<v Speaker 2>need to be clear with that last one, because it's

0:13:54.200 --> 0:13:57.320
<v Speaker 2>a direct quote from the information hm M. The company

0:13:57.320 --> 0:13:59.880
<v Speaker 2>also expects growth and inference costs the costs of running

0:13:59.880 --> 0:14:02.640
<v Speaker 2>A products such as chat GPT in their underlying models

0:14:02.679 --> 0:14:05.320
<v Speaker 2>to moderate over the next half decade. These costs will

0:14:05.320 --> 0:14:07.760
<v Speaker 2>triple this year referring to twenty twenty five to six

0:14:07.800 --> 0:14:11.600
<v Speaker 2>billion dollars, and rise to nearly forty seven billion dollars

0:14:11.679 --> 0:14:14.800
<v Speaker 2>in twenty thirty. Still, the annual growth rate will fall

0:14:14.800 --> 0:14:18.120
<v Speaker 2>to about thirty percent. Then, okay, thanks Also, are you

0:14:18.200 --> 0:14:21.960
<v Speaker 2>fucking kidding me? Six billion fucking dollars for fucking influence? Hey,

0:14:22.000 --> 0:14:24.440
<v Speaker 2>ca C Newton, I thought those costs were coming down.

0:14:24.600 --> 0:14:32.560
<v Speaker 2>KC case, KC. Wow, he's not here. He's not here anyway.

0:14:32.640 --> 0:14:35.480
<v Speaker 2>That's not really great at all. That's actually really bad.

0:14:35.720 --> 0:14:38.200
<v Speaker 2>The Information reports that open ai will make about eight

0:14:38.200 --> 0:14:41.560
<v Speaker 2>billion dollars some subscriptions to chat GPT in twenty twenty five,

0:14:41.760 --> 0:14:44.680
<v Speaker 2>meaning that seventy five percent of open AI's largest revenue

0:14:44.720 --> 0:14:46.720
<v Speaker 2>source is eaten up by the price of providing it.

0:14:47.520 --> 0:14:49.480
<v Speaker 2>This is meant to be the cheap part. This is

0:14:49.520 --> 0:14:52.000
<v Speaker 2>the one fucking thing people say to me is meant

0:14:52.000 --> 0:14:54.080
<v Speaker 2>to come down in price. I've had assholes saying to

0:14:54.120 --> 0:14:56.440
<v Speaker 2>me for the last year. Custom influence is coming down?

0:14:56.800 --> 0:14:59.680
<v Speaker 2>Is it? Are we living in different dimensions? Are there

0:14:59.760 --> 0:15:01.560
<v Speaker 2>large as parts of the tech media that have fucking

0:15:01.600 --> 0:15:04.640
<v Speaker 2>gas leaks? What am I missing? Tell me what I

0:15:04.680 --> 0:15:06.960
<v Speaker 2>am missing near ed. You haven't take to people to

0:15:07.040 --> 0:15:09.200
<v Speaker 2>Billy Sea. You don't need it it. Shut the fuck up.

0:15:09.240 --> 0:15:10.800
<v Speaker 2>If you are one of these people who says I

0:15:10.880 --> 0:15:13.720
<v Speaker 2>need to in Casey, you're included, man. Fuck like, I'm

0:15:13.760 --> 0:15:15.960
<v Speaker 2>so sick of this, Oh you don't talk to people

0:15:16.080 --> 0:15:19.040
<v Speaker 2>running these things. I am sick of people like Casey,

0:15:19.120 --> 0:15:21.880
<v Speaker 2>you and and others too saying, Oh, you don't talk

0:15:21.920 --> 0:15:24.600
<v Speaker 2>to enough AI people. You haven't listened to them. You mean,

0:15:24.680 --> 0:15:26.760
<v Speaker 2>I haven't listened to the problem of the people that

0:15:26.840 --> 0:15:30.320
<v Speaker 2>make money off of lying about this dog shit? Are

0:15:30.360 --> 0:15:33.720
<v Speaker 2>you really thinking you think that's what's missing from my analysis,

0:15:33.840 --> 0:15:36.160
<v Speaker 2>interviewing people who work at these companies and understanding how

0:15:36.200 --> 0:15:38.800
<v Speaker 2>the technologies work. I know other technologies work. I don't

0:15:38.800 --> 0:15:41.200
<v Speaker 2>need to talk to these fucking people. There are people

0:15:41.200 --> 0:15:43.520
<v Speaker 2>out there like Simon Wilson and Max Wolf who know

0:15:43.560 --> 0:15:45.760
<v Speaker 2>how these things work that I talk to fairly regularly,

0:15:45.800 --> 0:15:47.520
<v Speaker 2>and both of them push back on me because they

0:15:47.760 --> 0:15:51.200
<v Speaker 2>know how large language models work. Those people matter. What

0:15:51.360 --> 0:15:54.120
<v Speaker 2>doesn't matter to me, what will never matter to me

0:15:54.360 --> 0:15:56.800
<v Speaker 2>is what Dario, Emma, Dave, Jack Clark and all the

0:15:56.800 --> 0:15:59.760
<v Speaker 2>other fucking people anthropic thing and I think it's the

0:15:59.800 --> 0:16:05.000
<v Speaker 2>tech and actively honestly malpractice in journalism to pretend that

0:16:05.040 --> 0:16:08.720
<v Speaker 2>there's something ethical about speaking to these people and listening

0:16:08.800 --> 0:16:14.200
<v Speaker 2>and taking in their marketing spiel. It's actually a little

0:16:14.240 --> 0:16:18.840
<v Speaker 2>bit disgusting that this is even a critique leveled at anyone.

0:16:20.440 --> 0:16:21.720
<v Speaker 2>But you can have to forgive me. I'm gonna be

0:16:21.760 --> 0:16:23.400
<v Speaker 2>a little rude, and I know that seemed like it,

0:16:23.480 --> 0:16:25.920
<v Speaker 2>but I'm not even getting excited. In fact, you know what,

0:16:27.640 --> 0:16:31.560
<v Speaker 2>I think it's time, Okay, everyone, I think it's time

0:16:32.160 --> 0:16:34.640
<v Speaker 2>that I go through the most common critiques in AI.

0:16:34.720 --> 0:16:37.400
<v Speaker 2>It's time for me to really sit down. And I'm

0:16:37.400 --> 0:16:39.120
<v Speaker 2>going to do my Kevin Rooth's voice. And I know

0:16:39.280 --> 0:16:41.360
<v Speaker 2>a lot of you like my Kevin Ruth's voice, and

0:16:41.480 --> 0:16:43.680
<v Speaker 2>some of you, not a lot of you. I'm going

0:16:43.720 --> 0:16:45.160
<v Speaker 2>to say I'm being rude to these people, and it

0:16:45.160 --> 0:16:47.600
<v Speaker 2>weakens my analysis, to which I say, kiss my ass,

0:16:48.440 --> 0:16:50.480
<v Speaker 2>I will I will turn you, I will cube you

0:16:50.920 --> 0:16:55.200
<v Speaker 2>like a car in a garbage dump. But let's start,

0:16:55.320 --> 0:16:58.400
<v Speaker 2>shall we. The cost of inference coming down? That's one argument. Okay,

0:16:58.600 --> 0:17:02.840
<v Speaker 2>source sauce, where is your source? If you are someone

0:17:02.960 --> 0:17:05.760
<v Speaker 2>saying to me that the costs of inference are coming down.

0:17:06.320 --> 0:17:08.159
<v Speaker 2>I want your source. I want you to show me

0:17:08.240 --> 0:17:09.840
<v Speaker 2>the costs. I want you to show me the costs

0:17:09.920 --> 0:17:12.600
<v Speaker 2>at scale, because it sure seems like they're increasing for

0:17:12.680 --> 0:17:15.240
<v Speaker 2>open ai, and they're effectively the entire user base of

0:17:15.280 --> 0:17:18.560
<v Speaker 2>the entire generative AI industry. But Edward about deep U,

0:17:18.760 --> 0:17:21.359
<v Speaker 2>sweet idiot child. Deep Seek is not open Ai, and

0:17:21.400 --> 0:17:23.920
<v Speaker 2>open AI's latest models only seem to be getting more

0:17:23.920 --> 0:17:26.920
<v Speaker 2>expensive as time drags on. GPT four point five costs

0:17:26.920 --> 0:17:29.639
<v Speaker 2>seventy five dollars per million input tokens and one hundred

0:17:29.680 --> 0:17:32.080
<v Speaker 2>and fifty dollars per million output tokens, and that the

0:17:32.160 --> 0:17:34.880
<v Speaker 2>risk of repeating myself. Open ai is effectively the entire

0:17:34.960 --> 0:17:37.800
<v Speaker 2>generitive AI industry, at least for the world outside of China.

0:17:37.840 --> 0:17:41.760
<v Speaker 2>On top of that, we actually don't know whether deep

0:17:41.760 --> 0:17:44.560
<v Speaker 2>seak is even profitable to run at scale. It is

0:17:44.640 --> 0:17:48.240
<v Speaker 2>definitely cheaper to run, but we don't know if it's

0:17:48.240 --> 0:17:51.320
<v Speaker 2>actually profitable. Indeed, I don't know, even though how you

0:17:51.520 --> 0:17:54.520
<v Speaker 2>calculate this, because running a deep seek model is just

0:17:54.600 --> 0:17:57.040
<v Speaker 2>one person is one thing. The question is whether you

0:17:57.080 --> 0:17:59.840
<v Speaker 2>could scale it up like open ai. We don't know.

0:18:00.000 --> 0:18:02.399
<v Speaker 2>But let's get up back to the other critiques. This

0:18:02.520 --> 0:18:04.680
<v Speaker 2>is a company, it's growth stage. They can just hit

0:18:04.720 --> 0:18:07.240
<v Speaker 2>the button, it all be profitable. You have the mind

0:18:07.240 --> 0:18:09.560
<v Speaker 2>of a child. If this was the case, why would

0:18:09.560 --> 0:18:12.159
<v Speaker 2>both anthropic and open AI be losing so much money?

0:18:12.280 --> 0:18:14.800
<v Speaker 2>Why are none of the hyper scalers making profit on AI?

0:18:14.920 --> 0:18:17.480
<v Speaker 2>Why does nobody want to talk about the underlying economics

0:18:17.680 --> 0:18:20.240
<v Speaker 2>if they're at the growth stage? And also a little

0:18:20.280 --> 0:18:22.200
<v Speaker 2>side point as well, why have we been at the

0:18:22.200 --> 0:18:24.880
<v Speaker 2>growth stage for years? And why are hyper scalers at

0:18:24.880 --> 0:18:28.720
<v Speaker 2>the growth stage? They're not startups anyway, on to another one, though,

0:18:28.760 --> 0:18:30.840
<v Speaker 2>these are the early days of AI. It's just that

0:18:30.920 --> 0:18:34.680
<v Speaker 2>the early days wrong. Wrong. We have all the king's

0:18:34.720 --> 0:18:36.960
<v Speaker 2>horses and all the king's men, the entire tech industry,

0:18:37.119 --> 0:18:39.600
<v Speaker 2>and more money that has ever been invested into anything

0:18:39.640 --> 0:18:42.960
<v Speaker 2>piled into generative AI, and the result has been utterly mediocre.

0:18:43.359 --> 0:18:46.959
<v Speaker 2>Nobody's making on money on AI other than Nvidia and

0:18:47.000 --> 0:18:51.080
<v Speaker 2>maybe during a consultancy, but ed they're already showing signs

0:18:51.119 --> 0:18:54.560
<v Speaker 2>that the AI is going to be powerful. No it's not. No,

0:18:54.680 --> 0:18:57.320
<v Speaker 2>it's not like I'm If anyone brings these critiques, you

0:18:57.440 --> 0:19:00.399
<v Speaker 2>just say no, no, they're not. Show me, show me,

0:19:01.000 --> 0:19:04.240
<v Speaker 2>show me, why is it the only people I'm giving

0:19:04.240 --> 0:19:06.640
<v Speaker 2>Simon Wilson credit here. He's one of the only people

0:19:06.680 --> 0:19:08.960
<v Speaker 2>who'll show you anything cool. And it's cloud compute stuff.

0:19:09.000 --> 0:19:11.879
<v Speaker 2>It's like relatively boring enterprise stuff. It's exciting for the

0:19:12.000 --> 0:19:16.480
<v Speaker 2>niche cases, like software generally is, but it's really not

0:19:16.680 --> 0:19:19.680
<v Speaker 2>showing any power. We talk about this powerful AI thing,

0:19:19.840 --> 0:19:22.840
<v Speaker 2>is it in the room? Like? Where is this? Where

0:19:22.880 --> 0:19:25.719
<v Speaker 2>is this powerful AI? But then I have actually had

0:19:25.720 --> 0:19:28.280
<v Speaker 2>a few emails saying, ed ed look at Open Eyes

0:19:28.480 --> 0:19:31.040
<v Speaker 2>three model, and I just want to be clear that

0:19:31.119 --> 0:19:35.760
<v Speaker 2>this new and extremely expensive reasoning model also hallucinates more.

0:19:36.800 --> 0:19:40.399
<v Speaker 2>Is that AGI? By the way, is this AGI? Is

0:19:40.480 --> 0:19:43.040
<v Speaker 2>the AGI in the room with us? Did the AGI

0:19:43.200 --> 0:19:45.119
<v Speaker 2>tell you it loved you? Did it tell you to

0:19:45.200 --> 0:19:48.760
<v Speaker 2>leave your wife? Did it offer you sex? I hope you're.

0:19:48.640 --> 0:20:07.119
<v Speaker 1>Okay, But ed Edie really is the early days though

0:20:07.119 --> 0:20:08.760
<v Speaker 1>it's just like this in the early days of the Internet.

0:20:08.840 --> 0:20:13.880
<v Speaker 2>No, it was not, and you're a buffoon for suggesting otherwise.

0:20:14.280 --> 0:20:16.439
<v Speaker 2>Jim Cavello of Goldman Sachs wrote in a note from

0:20:16.520 --> 0:20:18.640
<v Speaker 2>last year and the episode pop Culture, you can listen

0:20:18.680 --> 0:20:21.600
<v Speaker 2>to it, which the early days of the Internet were

0:20:21.640 --> 0:20:24.520
<v Speaker 2>nothing like this. Nothing at all. Nothing. There were these

0:20:24.560 --> 0:20:28.240
<v Speaker 2>sixty two thousand and sixty four thousand dollars some microsystems, servers, yes,

0:20:28.280 --> 0:20:32.000
<v Speaker 2>but there were so so many few of them. But

0:20:32.280 --> 0:20:35.199
<v Speaker 2>ed smartphones, I've got you. I finally have me and

0:20:35.680 --> 0:20:38.040
<v Speaker 2>my sex. People doubted those two, they didn't. I will

0:20:38.119 --> 0:20:41.639
<v Speaker 2>drown you in an icy lake if another person comes

0:20:41.680 --> 0:20:46.040
<v Speaker 2>to me and says, hey, ed smartphones. People doubted smartphones.

0:20:46.119 --> 0:20:48.720
<v Speaker 2>Nobody darted smartphones. Why do people get I've read this

0:20:48.760 --> 0:20:50.679
<v Speaker 2>point so many times, but no one seems to have

0:20:50.680 --> 0:20:54.960
<v Speaker 2>a fucking hyperlink because they're lying. They goddamn lying. Cavello

0:20:55.040 --> 0:20:58.320
<v Speaker 2>of Goldman Sachs also noted and including an entire thing

0:20:58.320 --> 0:21:01.040
<v Speaker 2>about how smartphones were fully telegram after analysts and Advance

0:21:01.080 --> 0:21:04.960
<v Speaker 2>with hundreds of presentations that accurately fit house smartphones rolled out,

0:21:05.359 --> 0:21:09.400
<v Speaker 2>and that said that no roadmap exists for AI. It's

0:21:09.520 --> 0:21:13.760
<v Speaker 2>just we're years into this and I'm still repeating the

0:21:13.760 --> 0:21:16.359
<v Speaker 2>same points, and I still don't have much in return

0:21:16.440 --> 0:21:18.480
<v Speaker 2>other than the cost of inference that going down. But

0:21:19.440 --> 0:21:22.040
<v Speaker 2>here's another point of people like it. They go out ed,

0:21:22.080 --> 0:21:23.800
<v Speaker 2>you're so bone and check out this article, and some

0:21:23.880 --> 0:21:26.000
<v Speaker 2>of you love to emailed me this fucking thing. Not

0:21:26.080 --> 0:21:28.640
<v Speaker 2>many of you, I must be clear. The listeners. You're wonderful.

0:21:28.680 --> 0:21:30.240
<v Speaker 2>I love you so much, but there's one or two

0:21:30.280 --> 0:21:32.760
<v Speaker 2>of you out there. Really you're very attached to your

0:21:32.800 --> 0:21:34.760
<v Speaker 2>generative ais and I'm never gonna like it. But some

0:21:34.840 --> 0:21:37.320
<v Speaker 2>of you like to send me this article from Newsweek

0:21:37.520 --> 0:21:39.919
<v Speaker 2>in nineteen ninety five from a guy who said that

0:21:39.960 --> 0:21:42.160
<v Speaker 2>the web would not be a big business Clifford style.

0:21:42.200 --> 0:21:46.600
<v Speaker 2>He said, why the Web won't be Nirvana? And this piece,

0:21:46.680 --> 0:21:48.919
<v Speaker 2>by the way, is quite detailed. You should read it.

0:21:48.960 --> 0:21:51.080
<v Speaker 2>I'm going to have it in the episode notes. But

0:21:51.560 --> 0:21:55.760
<v Speaker 2>they think that sending me this that one guy one

0:21:55.800 --> 0:22:00.200
<v Speaker 2>guy was wrong. Once one guy he said that the

0:22:00.280 --> 0:22:03.240
<v Speaker 2>Internet wonn't be big, And this proves that I add

0:22:03.359 --> 0:22:07.320
<v Speaker 2>ze Tron what's that ninety nine us like twenty years

0:22:07.400 --> 0:22:12.880
<v Speaker 2>later because one guy said that the Internet won't be big,

0:22:13.000 --> 0:22:15.480
<v Speaker 2>that I am wrong? Summer motherfucker? Have you read the piece?

0:22:15.680 --> 0:22:17.679
<v Speaker 2>That's actually the thing. All of these are things that

0:22:17.720 --> 0:22:19.440
<v Speaker 2>you can box up and use. Some people who use

0:22:19.480 --> 0:22:22.320
<v Speaker 2>this half fast bullshit Clifford Stole basically says that the

0:22:22.440 --> 0:22:24.600
<v Speaker 2>Internet at the time was pretty limited, and yes, he

0:22:24.600 --> 0:22:26.320
<v Speaker 2>conflated that with the idea that he won't be big

0:22:26.320 --> 0:22:29.640
<v Speaker 2>in the future. However, Stole's piece, also, as Michael Hiltzier

0:22:29.720 --> 0:22:32.880
<v Speaker 2>wrote for The La Times, was alarmingly accurate about misinformation

0:22:33.160 --> 0:22:36.920
<v Speaker 2>and sleazy companies selling computerized replacements for education. In any case,

0:22:36.960 --> 0:22:38.800
<v Speaker 2>one guy saying that the Internet won't be big doesn't

0:22:38.840 --> 0:22:40.879
<v Speaker 2>make the fucking thing about Jennity, I v ai and

0:22:40.880 --> 0:22:43.320
<v Speaker 2>you were a simpleton if you think it does. One

0:22:43.320 --> 0:22:45.440
<v Speaker 2>guy being wrong in some way is not a response

0:22:45.440 --> 0:22:48.439
<v Speaker 2>to a criticism. I will crush you like a bug.

0:22:48.480 --> 0:22:51.520
<v Speaker 2>If this is your logic, I will eat you. I

0:22:51.600 --> 0:22:53.480
<v Speaker 2>will put you in my mouth like Kirby, and I'll

0:22:53.480 --> 0:22:55.000
<v Speaker 2>shit you out, and I will have the powers of

0:22:55.000 --> 0:22:58.560
<v Speaker 2>a dunce. Stole's analysis also isn't based on hundreds of

0:22:58.640 --> 0:23:01.520
<v Speaker 2>hours of research and endless reporting. Mine is I will

0:23:01.520 --> 0:23:04.439
<v Speaker 2>grab you from the ceiling like the war Master from Zelda,

0:23:04.720 --> 0:23:09.840
<v Speaker 2>and you will never be heard from again. Anyway. Another argument,

0:23:10.680 --> 0:23:12.639
<v Speaker 2>another argument that people are to give me is the

0:23:12.640 --> 0:23:15.639
<v Speaker 2>open ai and anthropic of research entities not business, and

0:23:15.640 --> 0:23:18.560
<v Speaker 2>that they are not focused on profit. Okay, so just

0:23:18.600 --> 0:23:20.399
<v Speaker 2>so we're clear that if that's the case, they're just

0:23:20.400 --> 0:23:23.600
<v Speaker 2>going to burn money forever. Is that the case or

0:23:23.600 --> 0:23:26.400
<v Speaker 2>are they going to hit like the be profitable button sometime. Also,

0:23:26.440 --> 0:23:28.560
<v Speaker 2>if open ai was a research entity, why does it

0:23:28.600 --> 0:23:30.640
<v Speaker 2>need forty billion dollars some soft bank or to change

0:23:30.640 --> 0:23:33.920
<v Speaker 2>its weird corporate structure to become a full profit Actually wait,

0:23:34.040 --> 0:23:37.560
<v Speaker 2>wait a second, that just occurred to me. Open ai

0:23:37.760 --> 0:23:40.800
<v Speaker 2>is as many as eight hundred million weekly active users.

0:23:41.680 --> 0:23:43.800
<v Speaker 2>That's proof of adoption, right, That's going to be an

0:23:43.840 --> 0:23:46.520
<v Speaker 2>argument that people have. There's some bloke on Blue Sky

0:23:47.000 --> 0:23:49.240
<v Speaker 2>who has just been responding to me every few days

0:23:49.440 --> 0:23:51.399
<v Speaker 2>with this kind of argument, saying, look, look at all

0:23:51.440 --> 0:23:54.920
<v Speaker 2>the users, and look, I get the look. You might

0:23:54.960 --> 0:23:57.200
<v Speaker 2>be a bit horny about this number, but something don't

0:23:57.200 --> 0:24:00.000
<v Speaker 2>make no sense about this number. Or March the first

0:24:00.000 --> 0:24:02.240
<v Speaker 2>twenty twenty five, open ai said that it had five

0:24:02.320 --> 0:24:05.320
<v Speaker 2>hundred million people who used chat GPT every week. Two

0:24:05.359 --> 0:24:07.960
<v Speaker 2>weeks later, Sam Mortman claimed that something like ten percent

0:24:08.000 --> 0:24:10.560
<v Speaker 2>of the world uses their systems a lot. They're referring

0:24:10.560 --> 0:24:13.160
<v Speaker 2>to chat GPTs, and the media took this to mean

0:24:13.160 --> 0:24:16.119
<v Speaker 2>that chat GPT is eight hundred million weekly active users.

0:24:16.440 --> 0:24:18.040
<v Speaker 2>I just want to be clear about something as well.

0:24:18.040 --> 0:24:21.480
<v Speaker 2>Sam Moltman didn't say that, he said the weird vague

0:24:21.520 --> 0:24:24.560
<v Speaker 2>thing about something like ten percent of the world, like

0:24:24.640 --> 0:24:26.760
<v Speaker 2>that's what he said, and everyone just went, oh shit,

0:24:26.760 --> 0:24:29.120
<v Speaker 2>we gotta help help Sam Moorman out got to push

0:24:29.119 --> 0:24:31.760
<v Speaker 2>this bad boy over the edge. And there are three

0:24:31.760 --> 0:24:34.040
<v Speaker 2>ways to interpret what he said, and you tell me

0:24:34.080 --> 0:24:36.920
<v Speaker 2>which one sounds real. Number One, open AI's user base

0:24:36.960 --> 0:24:39.840
<v Speaker 2>increased by three hundred million weekly active users in two weeks.

0:24:40.880 --> 0:24:44.560
<v Speaker 2>Number two. Open ai understated it's user base in the

0:24:44.560 --> 0:24:47.520
<v Speaker 2>announcement of their funding announcement on OpenAI dot com by

0:24:47.520 --> 0:24:52.320
<v Speaker 2>three hundred million users or three. Number three, How about this,

0:24:52.600 --> 0:24:56.240
<v Speaker 2>Sam Moltman fucking lied. I get that some members of

0:24:56.320 --> 0:24:58.760
<v Speaker 2>the media of a weird attachment to this damp little man,

0:24:58.800 --> 0:25:01.119
<v Speaker 2>But have any of you ever considered that he's just

0:25:01.160 --> 0:25:03.280
<v Speaker 2>fucking saying things knowing that you'll print them with the

0:25:03.320 --> 0:25:07.399
<v Speaker 2>kindest possible interpretation. Sam Oltman is a liar. He's lied before,

0:25:07.400 --> 0:25:10.040
<v Speaker 2>and he'll lie again. I wrote an entire newsletter called

0:25:10.040 --> 0:25:11.960
<v Speaker 2>sam Oltman is Full of Shit. You should read it.

0:25:11.960 --> 0:25:14.560
<v Speaker 2>I'm gonna link to it, but way ed Google says

0:25:14.560 --> 0:25:16.720
<v Speaker 2>it as three hundred and fifty million monthly active uses

0:25:16.760 --> 0:25:20.359
<v Speaker 2>on Gemini eat shit Zichron No, you eat shit? Yes,

0:25:20.440 --> 0:25:23.600
<v Speaker 2>Google Gemini has three hundred and fifty million monthly active users,

0:25:23.600 --> 0:25:26.919
<v Speaker 2>and that's because they started replacing Google Assistant with Google

0:25:26.960 --> 0:25:30.080
<v Speaker 2>Gemini in early March. You are being had, You are

0:25:30.119 --> 0:25:33.720
<v Speaker 2>being swindled. If Google replaced Google Search with Google Gemini,

0:25:33.720 --> 0:25:38.080
<v Speaker 2>it would have billions of monthly active users. Jesus Christ,

0:25:38.840 --> 0:25:42.520
<v Speaker 2>Jesus Christ on a goddamn cracker. Even reading this script out,

0:25:42.520 --> 0:25:45.240
<v Speaker 2>I get like some of you have suggested that this

0:25:45.280 --> 0:25:48.960
<v Speaker 2>is at all manufactured. No. Reading this stuff makes me

0:25:49.080 --> 0:25:53.200
<v Speaker 2>very angry because I didn't grow up popular or intelligent

0:25:53.200 --> 0:25:55.040
<v Speaker 2>in any way. I've had to pick this shit up

0:25:55.200 --> 0:25:57.520
<v Speaker 2>as I go, and I don't think what I'm saying

0:25:57.640 --> 0:26:00.119
<v Speaker 2>is crazy, but I am sometimes treated that way, and

0:26:00.160 --> 0:26:03.800
<v Speaker 2>this episode I realize I'm doing myself no favors. But anyway,

0:26:03.800 --> 0:26:06.679
<v Speaker 2>back to the critics, really quickly open AI having hundreds

0:26:06.720 --> 0:26:08.960
<v Speaker 2>of millions of free years as each losing it money

0:26:09.000 --> 0:26:11.720
<v Speaker 2>is proof that the free version of chat GPT is popular,

0:26:12.000 --> 0:26:14.600
<v Speaker 2>largely because the entirety of the media has written about

0:26:14.640 --> 0:26:17.520
<v Speaker 2>AI NonStop for two straight years and mentioned chat GPT

0:26:17.680 --> 0:26:20.760
<v Speaker 2>every single fucking time. Yes, yes, there is a degree

0:26:20.800 --> 0:26:22.720
<v Speaker 2>here of marketing at partnerships of word of mouth of

0:26:22.840 --> 0:26:25.720
<v Speaker 2>some degree of utility. But when you remove the non

0:26:25.760 --> 0:26:29.440
<v Speaker 2>stop free media campaign chat GPT would have petered out

0:26:29.720 --> 0:26:32.480
<v Speaker 2>by now along with this stupid fucking bubble, but edits

0:26:32.480 --> 0:26:34.960
<v Speaker 2>pouf sambody's doing something. Yeah, it's proved that something is

0:26:34.960 --> 0:26:39.240
<v Speaker 2>broken in society. Generative AI has never ever had the

0:26:39.359 --> 0:26:43.119
<v Speaker 2>kind of meaningful business returns or utility that actually underpins

0:26:43.119 --> 0:26:46.080
<v Speaker 2>something meaningful, but it has had enough to make people

0:26:46.119 --> 0:26:50.080
<v Speaker 2>give it a try. Do you not? Actually? No, I

0:26:50.119 --> 0:26:52.840
<v Speaker 2>know you listening. You're gonna get this. In some ways

0:26:52.880 --> 0:26:54.880
<v Speaker 2>this episode has been I mean, in all ways it's

0:26:54.880 --> 0:26:57.399
<v Speaker 2>been pretty rant. In the second one going to be

0:26:57.440 --> 0:26:59.720
<v Speaker 2>even more so. What I'm trying to do here is

0:26:59.760 --> 0:27:03.000
<v Speaker 2>show you how fascicle all this crap is, how ridiculous

0:27:03.000 --> 0:27:07.240
<v Speaker 2>it is, how silly these posits are. These projections are

0:27:07.280 --> 0:27:10.080
<v Speaker 2>the suggestion that what we have today will become something else,

0:27:10.359 --> 0:27:13.439
<v Speaker 2>when all we've had is proof that it won't. Do

0:27:13.480 --> 0:27:16.600
<v Speaker 2>you see the obvious cracks in the wall here, No

0:27:16.640 --> 0:27:20.840
<v Speaker 2>matter how strenuously people like professional credulous dipshits that the

0:27:21.240 --> 0:27:24.320
<v Speaker 2>other big publications tried to pave over them, does any

0:27:24.320 --> 0:27:26.720
<v Speaker 2>of this make sense to you? Because I, even when

0:27:26.760 --> 0:27:29.040
<v Speaker 2>I try and steal man. My own arguments. Can't wrap

0:27:29.080 --> 0:27:31.520
<v Speaker 2>my head around how any of this survives, let alone

0:27:31.520 --> 0:27:33.840
<v Speaker 2>becomes an industry where the biggest player has annual revenue

0:27:33.880 --> 0:27:37.479
<v Speaker 2>is greater than some major industrialized countries. And I know

0:27:37.680 --> 0:27:40.439
<v Speaker 2>some of you the emotions a lot, and I know

0:27:40.480 --> 0:27:44.320
<v Speaker 2>the aggressions a lot. I'm frustrated because I truly believe

0:27:44.440 --> 0:27:47.480
<v Speaker 2>this stuff's falling apart. I truly believe that this was

0:27:47.760 --> 0:27:50.960
<v Speaker 2>never really anything well. I'm saying this. Kevin Rose is

0:27:51.000 --> 0:27:53.000
<v Speaker 2>in the New York Times going, I believe that Agi

0:27:53.080 --> 0:27:55.639
<v Speaker 2>is my friend. I believe AGI will rise out of

0:27:55.680 --> 0:27:57.560
<v Speaker 2>the ground and hug me in the way no one

0:27:57.640 --> 0:28:02.199
<v Speaker 2>ever has. I think that's disgusting on levels. But I

0:28:02.240 --> 0:28:06.120
<v Speaker 2>also think it's genuinely irresponsible. I think all of this

0:28:06.200 --> 0:28:10.119
<v Speaker 2>is I think when this collapses, we're going to have

0:28:10.200 --> 0:28:13.160
<v Speaker 2>to look back and take inventory of how we got here.

0:28:15.119 --> 0:28:17.480
<v Speaker 2>And I need you to in the next episode listen

0:28:17.560 --> 0:28:21.119
<v Speaker 2>to it through the kind of len listen to it

0:28:21.160 --> 0:28:23.919
<v Speaker 2>through the lens. That's how lenses work. I need you

0:28:23.960 --> 0:28:25.720
<v Speaker 2>to just stick with it and realize that all of

0:28:25.720 --> 0:28:28.040
<v Speaker 2>the what this is is trying to show you and

0:28:28.119 --> 0:28:31.000
<v Speaker 2>hopefully other people that you talk to how silly this is,

0:28:31.160 --> 0:28:33.159
<v Speaker 2>how ridiculous this is, and that we have a major

0:28:33.200 --> 0:28:36.119
<v Speaker 2>problem in tech and business media. We have a problem

0:28:36.160 --> 0:28:39.000
<v Speaker 2>where people can come out and just say whatever the

0:28:39.160 --> 0:28:43.680
<v Speaker 2>Charlie Brown had hose of the tech media. And it's

0:28:43.720 --> 0:28:46.520
<v Speaker 2>disgusting to me because there are startups that could use

0:28:46.520 --> 0:28:48.719
<v Speaker 2>this money. There are better things to be done with

0:28:48.760 --> 0:28:51.400
<v Speaker 2>this money. Perhaps they're not hypergrowth markets, but there are

0:28:51.400 --> 0:28:54.040
<v Speaker 2>things that actually exist that could be piled into instead.

0:28:54.080 --> 0:28:57.640
<v Speaker 2>We've done this to make companies look like they can grow,

0:28:57.880 --> 0:29:01.520
<v Speaker 2>to make Sam Altman able to buy another five million

0:29:01.600 --> 0:29:04.960
<v Speaker 2>dollar koncig car. Is that the one he has? Either way,

0:29:05.760 --> 0:29:07.959
<v Speaker 2>I'm not gonna lower the temperature on the next episode.

0:29:07.960 --> 0:29:10.000
<v Speaker 2>I'm gonna be honest. It's gonna be just the spicy.

0:29:10.640 --> 0:29:12.480
<v Speaker 2>But I want you to know all of this frustration

0:29:12.600 --> 0:29:14.400
<v Speaker 2>comes from a place of knowing that we can do better,

0:29:14.480 --> 0:29:17.400
<v Speaker 2>and knowing that the tech industry could do better. Perhaps

0:29:17.480 --> 0:29:19.200
<v Speaker 2>it won't be as big as it is today in

0:29:19.240 --> 0:29:22.400
<v Speaker 2>the future, I don't know, but for it to get better,

0:29:22.840 --> 0:29:26.080
<v Speaker 2>this shit needs to end. Stick around for the next part,

0:29:26.120 --> 0:29:28.400
<v Speaker 2>where I'll talk about how we actually got here, how

0:29:28.400 --> 0:29:31.240
<v Speaker 2>this bubble got inflated, and how nasty the result could

0:29:31.280 --> 0:29:42.080
<v Speaker 2>be At the end, thank you for listening to Better Offline.

0:29:42.200 --> 0:29:44.640
<v Speaker 2>The editor and composer of the Better Offline theme song

0:29:44.720 --> 0:29:47.360
<v Speaker 2>is Matasowski. You can check out more of his music

0:29:47.360 --> 0:29:50.880
<v Speaker 2>and audio projects at Matasowski dot com, M A T

0:29:51.040 --> 0:29:55.640
<v Speaker 2>T O. S O w Ski dot com. You can

0:29:55.680 --> 0:29:58.160
<v Speaker 2>email me at easy at Better offline dot com or

0:29:58.240 --> 0:30:00.720
<v Speaker 2>visit Better Offline dot com to find more podcast links

0:30:00.760 --> 0:30:03.920
<v Speaker 2>and of course, my newsletter. I also really recommend you

0:30:03.920 --> 0:30:06.080
<v Speaker 2>go to chat dot Where's youoead dot at to visit

0:30:06.120 --> 0:30:08.760
<v Speaker 2>the discord, and go to our slash Better Offline to

0:30:08.800 --> 0:30:11.920
<v Speaker 2>check out our reddit. Thank you so much for listening.

0:30:12.760 --> 0:30:15.440
<v Speaker 2>Better Offline is a production of cool Zone Media. For

0:30:15.560 --> 0:30:18.720
<v Speaker 2>more from cool Zone Media, visit our website cool Zonemedia

0:30:18.760 --> 0:30:21.640
<v Speaker 2>dot com, or check us out on the iHeartRadio app,

0:30:21.680 --> 0:30:24.360
<v Speaker 2>Apple Podcasts, or wherever you get your podcasts.