WEBVTT - Week in Tech: There's No Place Like AI 0:00:12.280 --> 0:00:15.200 Welcome to tech Stuff, a production of iHeart Podcasts and 0:00:15.280 --> 0:00:19.080 Kaleidoscope iMOS vloscan and carries out again this week. So 0:00:19.120 --> 0:00:23.200 today I'll bring you the headlines, including what Google Search 0:00:23.480 --> 0:00:27.360 sounds like and why you're out of stock Cereal may 0:00:27.360 --> 0:00:30.440 be the result of a cyber attack on today's tech 0:00:30.480 --> 0:00:36.320 support segment. Semaphor's read Albergotti casts AI companies in the 0:00:36.360 --> 0:00:37.159 Wizard of Oz. 0:00:37.840 --> 0:00:40.720 The funniest part of this whole thing, honestly, was reaching 0:00:40.760 --> 0:00:43.640 out to all of these companies and explaining to them 0:00:43.680 --> 0:00:47.400 why they are the Munchkins or the wicked Witch of 0:00:47.440 --> 0:00:48.600 the East or the West. 0:00:49.920 --> 0:00:53.400 All of that in the Weekend Tech. It's Friday, June twentieth. 0:00:55.760 --> 0:01:01.160 In summary, today's headlines are about Summary's perhaps the most 0:01:01.240 --> 0:01:04.840 ubiquitous artifact of the age of AI, and today we 0:01:04.880 --> 0:01:08.040 have two stories about AI summaries, one about a new 0:01:08.120 --> 0:01:11.680 AI feature on Google Search and the other about Wikipedia's 0:01:11.760 --> 0:01:15.720 turn away from AI. We'll start with Google. As you 0:01:15.800 --> 0:01:19.160 might know, Google has this product called Notebook LM, which 0:01:19.200 --> 0:01:24.319 allows users to feed their own research sources, PDFs, websites, spreadsheets, 0:01:24.360 --> 0:01:27.560 whatever it may be, and Notebook LM will analyze the 0:01:27.600 --> 0:01:30.160 data in all kinds of different ways. One of these 0:01:30.160 --> 0:01:33.840 ways is a quote audio overview, where users can generate 0:01:33.959 --> 0:01:37.120 this two hander deep dive podcast where a male voice 0:01:37.120 --> 0:01:40.080 and a female voice talk each other through whatever the 0:01:40.080 --> 0:01:43.759 topic at hand is. Now, Google is bringing this feature 0:01:44.000 --> 0:01:47.600 to its search results. According to Ours Technica, the feature 0:01:47.640 --> 0:01:50.440 hasn't been fully rolled out yet. Users have to opt 0:01:50.440 --> 0:01:53.320 in by going to labs dot Google dot com slash 0:01:53.360 --> 0:01:57.040 search and turning on audio overviews. There seems to be 0:01:57.120 --> 0:01:59.280 a wait list, but one of our producers was able 0:01:59.320 --> 0:02:02.640 to turn the future before news spread too far, and 0:02:02.880 --> 0:02:06.840 here's what she found in the search results. Underneath the 0:02:06.880 --> 0:02:10.720 section that says quote, people also ask, there's an oval 0:02:10.760 --> 0:02:15.840 button that says generate audio overview. When you click it, 0:02:15.840 --> 0:02:19.079 it starts working because even AI podcasts take a little 0:02:19.120 --> 0:02:22.040 bit of time to cook. So while you wait, Google 0:02:22.080 --> 0:02:24.679 tells you exactly what it's doing, which is quote, understanding 0:02:24.680 --> 0:02:29.840 the query intent, researching websites, generating transcript, and then in 0:02:29.919 --> 0:02:33.880 less than a minute, an audio player appears. So our 0:02:33.919 --> 0:02:37.840 producer searched in Google, what is two plus two? The 0:02:37.880 --> 0:02:41.760 resulting audio overview was two and a half minutes long, 0:02:42.200 --> 0:02:46.600 which already tickled her spidery senses. But before we listen 0:02:46.639 --> 0:02:49.120 to the audio, we checked out all the websites listed 0:02:49.200 --> 0:02:53.079 underneath the clip as sources. They were a Reddit thread 0:02:53.280 --> 0:02:56.320 asking why some people think two plus two equals five, 0:02:57.040 --> 0:03:00.639 a link to study dot com, a Wikipedia page for 0:03:00.919 --> 0:03:03.400 two plus two equals five, as well as a link 0:03:03.440 --> 0:03:06.760 to a TikTok meme around the question of two plus two, 0:03:07.320 --> 0:03:11.160 which I couldn't really understand, let and explain. And underneath 0:03:11.200 --> 0:03:17.000 all these links is another message quote generative AI is experimental. 0:03:17.880 --> 0:03:20.280 So here's the beginning of that two and a half 0:03:20.360 --> 0:03:23.639 minute audio overview for what's two plus two? 0:03:24.120 --> 0:03:24.240 So? 0:03:24.280 --> 0:03:26.760 What's two plus two? Seems like a no brainer, right four? 0:03:27.000 --> 0:03:27.600 End of story. 0:03:27.720 --> 0:03:29.400 Well, that's what we learn in grade school at least. 0:03:29.639 --> 0:03:32.239 But the idea that two plus two could equal something 0:03:32.240 --> 0:03:35.200 other than four pops up in some pretty interesting places, 0:03:35.360 --> 0:03:36.280 like where. 0:03:36.240 --> 0:03:38.440 I'm picturing some kind of crazy math equation. 0:03:38.600 --> 0:03:41.480 Now it's less about complex math and more about, shall 0:03:41.520 --> 0:03:43.240 we say, bending reality a little. 0:03:43.640 --> 0:03:46.440 The podcast has then launched into a discussion of instances 0:03:46.480 --> 0:03:49.000 where some people might say two plus two equals five, 0:03:49.600 --> 0:03:53.120 like in All Wells dystopian novel nineteen eighty four, in 0:03:53.120 --> 0:03:56.720 which he ponders a totalitarian government that might insist two 0:03:56.760 --> 0:03:59.880 plus two equals five. It also includes a joke about 0:03:59.920 --> 0:04:03.160 it mathematician, an engineer, and a lawyer answering two plus 0:04:03.200 --> 0:04:06.880 two questions two plus two will always before that. The 0:04:06.920 --> 0:04:11.320 AI podcast hosts had this to say in conclusion, it's. 0:04:11.240 --> 0:04:12.960 Kind of mind bending when you think about it. This 0:04:13.160 --> 0:04:16.160 simple little equation can represent so much more than just 0:04:16.360 --> 0:04:17.080 basic addition. 0:04:17.279 --> 0:04:20.000 It really does. It's a reminder to question things, to 0:04:20.040 --> 0:04:23.480 consider different viewpoints, and to be aware of how easily 0:04:23.720 --> 0:04:26.279 fax can be twisted or manipulated. 0:04:26.640 --> 0:04:29.360 My personal view is that the notebook LM product can 0:04:29.400 --> 0:04:31.400 be pretty great if you want to bring a bunch 0:04:31.400 --> 0:04:33.560 of your own narrowly bounded set of data or research 0:04:33.560 --> 0:04:35.560 into one place and then here it played back to 0:04:35.600 --> 0:04:38.960 you in podcast form. More power to you. The last 0:04:38.960 --> 0:04:41.840 time I use it myself was over the holidays when 0:04:41.920 --> 0:04:44.560 I was going to go and see the Nutcracker Ballet, 0:04:45.240 --> 0:04:47.719 and I kind of wanted to know what to expect 0:04:47.760 --> 0:04:49.720 so that I'd be able to enjoy it more. And 0:04:49.839 --> 0:04:53.080 I've fed Notebook LM the Wikipedia page about the Nutcracker, 0:04:53.520 --> 0:04:57.240 some reviews, some other information, and it summarized all of 0:04:57.279 --> 0:04:59.640 that and turn it into a ten minute podcast which 0:04:59.640 --> 0:05:01.920 I listen to on the way to the ballet, and 0:05:01.960 --> 0:05:05.080 I honestly think that gave me a better enjoyment of 0:05:05.080 --> 0:05:07.680 the ballet because I knew the context, but I wouldn't 0:05:07.680 --> 0:05:09.960 have actually done the work myself of reading all the materials. 0:05:10.000 --> 0:05:13.160 So that was kind of a cool use case. That said, 0:05:13.440 --> 0:05:15.719 I don't think when I search for something like what's 0:05:15.760 --> 0:05:20.280 two plus two, I need a audio conversational summary of 0:05:20.640 --> 0:05:23.560 the randomness of the Internet. It just doesn't really seem 0:05:24.000 --> 0:05:26.960 very relevant, and it seems more than anything like a 0:05:27.120 --> 0:05:30.560 solution in search of a problem, which brings us to 0:05:30.600 --> 0:05:33.920 our next story. There's another Internet titan that tried out 0:05:33.920 --> 0:05:39.040 an AI feature recently. Wikipedia. On June second, the organization 0:05:39.120 --> 0:05:44.680 that hosts and maintains Wikipedia, called the Wikimedia Foundation, announced 0:05:44.720 --> 0:05:48.120 it would run an experiment with AI. The project, called 0:05:48.240 --> 0:05:52.240 Simple Article Summaries, would provide AI writeups at the top 0:05:52.400 --> 0:05:55.400 of Wikipedia entries. It was an attempt to make the 0:05:55.440 --> 0:05:59.279 denser Wikipedia articles more accessible, and the experiment was set 0:05:59.279 --> 0:06:02.360 to last two weeks, with summaries generated by an NLM 0:06:02.400 --> 0:06:08.320 called Aya by Coheer. Then came the criticism. According to 0:06:08.480 --> 0:06:11.240 a form media, the announcement was met with replies from 0:06:11.240 --> 0:06:14.520 Wikipedia editors who warned that AI summaries were a very 0:06:14.560 --> 0:06:18.040 bad idea. One editor pointed out that AI's ability to 0:06:18.080 --> 0:06:21.919 get things wrong could damage the platform's reputation, which is 0:06:21.960 --> 0:06:24.520 something that editors and the organization have worked very hard 0:06:24.560 --> 0:06:27.560 on for a very long time. Another editor said quote, 0:06:28.000 --> 0:06:31.159 just because Google has rolled out it's AI summaries doesn't 0:06:31.160 --> 0:06:33.760 mean we need to one up them. Some even threatened 0:06:33.760 --> 0:06:37.160 to quit if Wikipedia didn't roll back the summaries. And 0:06:37.600 --> 0:06:41.200 here's the thing. Wikipedia relies on its editor's passion for 0:06:41.240 --> 0:06:46.360 the website to keep this participatory encyclopedia running. So after 0:06:46.360 --> 0:06:48.920 the editor upraw and just a day after the announcement 0:06:48.960 --> 0:06:52.880 of the experiment, Wikimedia said they would pause it, but 0:06:52.960 --> 0:06:56.080 that they were also still interested in AI summaries. It's 0:06:56.080 --> 0:06:58.640 not clear when they might try them out again, but 0:06:58.680 --> 0:07:02.480 it is clear how human editors feel about them. I'm 0:07:02.520 --> 0:07:05.839 not somebody who's implacably opposed to AI summaries. In fact, 0:07:06.040 --> 0:07:08.600 I use them myself. But what was really important to 0:07:08.600 --> 0:07:11.800 remember is that AI summaries rely on the input of 0:07:11.920 --> 0:07:15.040 human work and ingenuity, and with that in mind, I 0:07:15.080 --> 0:07:17.200 read this article in the financial time that I wanted 0:07:17.200 --> 0:07:20.840 to share. It had the headline AI alone cannot solve 0:07:20.880 --> 0:07:23.960 the productivity problem, and it talks about something I had 0:07:23.960 --> 0:07:28.000 no idea about, which is the average scientist today produces 0:07:28.280 --> 0:07:32.120 fewer breakthrough ideas per dollar spent than their counterpart in 0:07:32.120 --> 0:07:35.440 the sixties, and that despite all the headlines about progress, 0:07:35.760 --> 0:07:39.960 progress is actually becoming more incremental. And here's the part 0:07:39.960 --> 0:07:43.320 of the article that really got me thinking. Quote, had 0:07:43.320 --> 0:07:47.040 the nineteenth century focused solely on better looms and plows, 0:07:47.400 --> 0:07:50.920 we would enjoy cheap cloth and abundant grain, but there 0:07:50.960 --> 0:07:55.440 would be no antibiotics, jet engines, or rockets. Economic miracles 0:07:55.480 --> 0:08:00.200 stem from discovery, not repeating tasks at greater speed. The 0:08:00.280 --> 0:08:02.360 article went on to make the point that large language 0:08:02.400 --> 0:08:09.280 models gravitate towards statistical consensus. So think about it. If 0:08:09.320 --> 0:08:12.240 there were NLMs back before the Wright Brothers, all the 0:08:12.320 --> 0:08:14.880 data that the model was trained on would have suggested 0:08:14.880 --> 0:08:19.480 that human flight was absolutely impossible. Wikipedia is not going 0:08:19.520 --> 0:08:23.160 to solve the larger problem of fostering human ingenuity in 0:08:23.240 --> 0:08:26.960 an age of statistical consensus, But at least for now, 0:08:27.160 --> 0:08:29.160 you can rest assured that the articles you read on 0:08:29.200 --> 0:08:40.840 it will be written and driven and created by humans. 0:08:41.880 --> 0:08:44.880 I've got a couple more headlines for you, including why 0:08:44.920 --> 0:08:47.520 you might be struggling to get your favorite box of cereal. 0:08:48.559 --> 0:08:51.760 Tech Crunch reports that some grocery stores are reeling from 0:08:51.760 --> 0:08:56.760 the effects of a cyber attack. United Natural Foods or UNFI, 0:08:57.400 --> 0:09:00.800 is a large food distribution company that provide it's fresh 0:09:00.840 --> 0:09:04.760 produce and products to over thirty thousand stores across the US. 0:09:05.200 --> 0:09:09.320 Whole Foods uses the company as its primary distributor. Earlier 0:09:09.320 --> 0:09:13.040 this month, UNFI experienced a cyber attack that caused them 0:09:13.080 --> 0:09:16.679 to shut down their electronic ordering systems. The system is 0:09:16.720 --> 0:09:19.920 now back online after more than ten days, but grocery 0:09:19.960 --> 0:09:22.840 stores are still waiting on some of their stock. And 0:09:22.880 --> 0:09:25.800 this isn't just an issue in the US. Cyber attacks 0:09:25.800 --> 0:09:30.360 are also roiling UK supermarkets. The thing is, grocery chains 0:09:30.440 --> 0:09:33.200 are a prime target for hackers because they have these large, 0:09:33.480 --> 0:09:37.920 interconnected supply chains, which by definition have multiple failure points. 0:09:38.360 --> 0:09:41.880 They also have treasure trows with customer data, and of course, 0:09:42.240 --> 0:09:44.360 when shoppers can't get what they want at the store, 0:09:44.880 --> 0:09:49.559 angst and headlines are sure to follow. Finally, for today's headlines, 0:09:50.160 --> 0:09:53.880 ads are a huge business for social media platforms, and 0:09:54.000 --> 0:09:56.520 TikTok is aiming to lower the cost of production for 0:09:56.559 --> 0:10:00.400 its partners. According to The Verge, TikTok is adding new 0:10:00.480 --> 0:10:04.280 capabilities to give advertisers the ability to upload images and 0:10:04.320 --> 0:10:06.959 give text prompts that can then be used to generate 0:10:07.160 --> 0:10:11.959 video advertisements. The videos can include virtual avatars holding or 0:10:12.040 --> 0:10:15.920 modeling different products, just like an influencer mine for brands 0:10:15.920 --> 0:10:18.600 that they partner with. TikTok has said that all this 0:10:18.760 --> 0:10:22.440 content generated using AI will be labeled as such and 0:10:22.520 --> 0:10:26.040 will go through quote multiple rounds of safety review. The 0:10:26.080 --> 0:10:29.920 biggest story is that influencer marketing transformed the ad industry 0:10:29.960 --> 0:10:32.400 over the last decade, and it will be interesting to 0:10:32.440 --> 0:10:36.640 see what AI influencers can actually do because, unlike say, 0:10:36.679 --> 0:10:39.600 fashion models, whose job it is to show what products 0:10:39.640 --> 0:10:43.160 will look like when they're worn, influencers have thrived when 0:10:43.200 --> 0:10:45.079 they make a case for a product based on their 0:10:45.120 --> 0:10:54.959 personal experience. After the break, we bring you a production 0:10:55.000 --> 0:10:58.440 of The Wizard of Oz starring your favorite tech giants. 0:10:58.880 --> 0:11:12.080 Stay with us on this show. We've talked about tech 0:11:12.160 --> 0:11:17.040 upstarts like open ai and its model, chatchipt Anthropic and 0:11:17.120 --> 0:11:20.920 its model, Clawed at Length, and We've also talked about 0:11:20.920 --> 0:11:25.480 the numerous tech giants racing to gain or defend market 0:11:25.480 --> 0:11:29.520 share with AI products of their own, including Meta, Google, Apple, Amazon, 0:11:29.600 --> 0:11:33.439 and Microsoft. It's a crowded field, but also an open 0:11:33.480 --> 0:11:36.640 one in terms of knowing who will ultimately prevail, and 0:11:36.720 --> 0:11:40.719 it can get quite dramatic their acquisitions, partnerships to get 0:11:40.760 --> 0:11:44.880 contentious legal fights and hedge rolling at big tech companies 0:11:44.920 --> 0:11:47.559 as they fight to stay in the game, and recently 0:11:47.679 --> 0:11:51.040 there have been some big AI power moves. Meta has 0:11:51.080 --> 0:11:55.000 just made its largest investments since acquiring WhatsApp, putting fourteen 0:11:55.040 --> 0:11:58.640 billion dollars into a company called Scale Ai. So we 0:11:58.640 --> 0:12:01.200 thought it'd be helpful to outline them many players the air, 0:12:01.240 --> 0:12:04.320 current positions, and what might be on the horizon. And 0:12:04.400 --> 0:12:06.760 here to walk us all through it is read Albagotti, 0:12:06.800 --> 0:12:09.960 who's the tech editor Sema fool Read Welcome to tech Stuff. 0:12:09.800 --> 0:12:11.160 Thanks for having me. Good to be here. 0:12:11.400 --> 0:12:15.199 You came up with a very creative paradigm for helping 0:12:15.720 --> 0:12:19.080 your readers make sense of all the different players in 0:12:19.200 --> 0:12:21.920 the AI race. Tell us a little bit about it. 0:12:22.760 --> 0:12:24.960 Well, it all started because, like you said, I mean, 0:12:25.000 --> 0:12:27.880 there's this this huge race, and I was sort of 0:12:27.920 --> 0:12:31.760 going back and forth with my colleague Rachel Jones, talking 0:12:31.760 --> 0:12:34.320 about Okay, so like, what are the lanes that these 0:12:34.360 --> 0:12:36.640 companies are starting to take and how do we sort 0:12:36.679 --> 0:12:39.160 of define that? And I'm sort of writing this thing 0:12:39.200 --> 0:12:41.319 and it's kind of it's a little dry, it's a 0:12:41.360 --> 0:12:44.520 little heavy on the analysis, and Rachel's like, why don't 0:12:44.559 --> 0:12:47.280 we just make them Wizard of Oz characters? And I 0:12:47.440 --> 0:12:53.120 was like, that's a terrible idea. But then I thought 0:12:53.120 --> 0:12:54.720 more about it, and I thought, that's actually kind of 0:12:54.720 --> 0:12:58.520 a great idea. So you know, we decided to try 0:12:58.520 --> 0:13:01.960 to fit you know, each of the biggest players that 0:13:02.000 --> 0:13:04.880 we see in the field into these characters and the 0:13:04.880 --> 0:13:08.480 Wizard of Oz and sort of explain why. And I think, look, 0:13:08.520 --> 0:13:11.000 I mean the funniest part of this whole thing, honestly, 0:13:11.280 --> 0:13:13.839 was was reaching out to all of these companies and 0:13:13.960 --> 0:13:17.000 explaining to them why they are the Munchkins or the 0:13:17.040 --> 0:13:21.720 Wicked Witch of the East or the West, and you know, 0:13:21.760 --> 0:13:24.679 getting their responses and I thought, you know, actually, kind 0:13:25.480 --> 0:13:27.480 I kind of like this because the best part was 0:13:27.520 --> 0:13:29.720 when I got pushed back from these companies, right like, 0:13:29.800 --> 0:13:32.559 we are we are not this, we are this character 0:13:32.840 --> 0:13:34.840 just like, oh, actually I'm getting an insight now into 0:13:34.880 --> 0:13:37.480 like how you see yourself as a company. And so 0:13:38.040 --> 0:13:39.439 what I kind of hoped was, like, you know, we 0:13:39.480 --> 0:13:42.079 would this was like people would write in and say 0:13:42.080 --> 0:13:45.160 you're a complete idiot, like open Aye is not Dorothy, 0:13:45.240 --> 0:13:47.360 They're you know, the tin man or something like that. 0:13:47.880 --> 0:13:50.640 Just get people talking about what are the lanes here 0:13:50.679 --> 0:13:53.440 and how and how should we think about these these players. 0:13:53.840 --> 0:13:56.200 But that said, you did cast open Aiye as Dorothy. 0:13:56.600 --> 0:13:57.199 Why was that? 0:13:57.800 --> 0:14:01.240 Well, you know, Dorothy is is kind of the main character. Right. 0:14:01.320 --> 0:14:04.720 They are the Kleenex of chat bots, so to speak. 0:14:04.920 --> 0:14:08.200 They're the winner take all sort of consumer chat bot. 0:14:08.640 --> 0:14:10.880 They get all the press, but you know, at the 0:14:10.920 --> 0:14:14.280 same time, they require a lot of other you know, 0:14:14.360 --> 0:14:16.800 players in this space to make it possible. Right, So 0:14:17.520 --> 0:14:21.520 you know they they've borrowed technology from Google, right, you know, 0:14:21.560 --> 0:14:25.200 they're using the data centers or at Microsoft for such 0:14:25.240 --> 0:14:27.800 a huge part of that. So it's this supporting cast 0:14:27.880 --> 0:14:31.320 that kind of really makes them who they are. I think, 0:14:31.840 --> 0:14:34.080 so that was that was the justification there. Do you 0:14:34.120 --> 0:14:36.080 agree or do you want to do you want to 0:14:36.080 --> 0:14:38.560 fight me on that? 0:14:38.680 --> 0:14:42.520 Well, I think I'll you've thought about this more than 0:14:42.520 --> 0:14:44.720 I have. But I think I agree with you that 0:14:45.560 --> 0:14:47.760 despite my name, I think that I agree with you 0:14:47.840 --> 0:14:50.200 that that that opening I definitely has the main character 0:14:50.320 --> 0:14:53.360 energy in this story. But I have to ask you 0:14:53.400 --> 0:14:56.760 about her cost of supporting characters the tin man, the 0:14:56.840 --> 0:14:59.320 cowardly line, and the scarecrow. How do you assign each 0:14:59.320 --> 0:15:00.720 of these crucial pots? 0:15:01.880 --> 0:15:05.720 Well, okay, so let's start with Let's start with with Google, 0:15:05.920 --> 0:15:09.720 the cowardly lion. Google has, you know, all this power, right, 0:15:09.760 --> 0:15:14.040 They've developed this technology for years. You know, the attention 0:15:14.240 --> 0:15:16.360 is all you need paper, right like that. That's what 0:15:16.520 --> 0:15:20.480 really laid the groundwork for CHATCHYPT. But they kind of 0:15:20.520 --> 0:15:22.440 sat on it. And I think a lot of that 0:15:22.600 --> 0:15:25.120 was out of fear. You know, these chatbots go off 0:15:25.120 --> 0:15:27.200 the rails. They're crazy, Like, what are people going to 0:15:27.280 --> 0:15:30.240 say if we just start releasing this stuff into the wild. 0:15:30.840 --> 0:15:34.680 And so they didn't and they got beat right, chatchapt 0:15:34.880 --> 0:15:38.880 came out. Google has had to completely reorient itself around 0:15:38.920 --> 0:15:39.840 this race now. 0:15:39.960 --> 0:15:42.480 So what you're saying is is Google could have had 0:15:42.520 --> 0:15:46.440 market leadership here, and because of fears about cultural blowback 0:15:46.560 --> 0:15:49.160 or erosion of that cool business, they basically sat on 0:15:49.200 --> 0:15:50.400 the technology. 0:15:49.920 --> 0:15:53.200 Yeah, that's very well said. They're a big public company. 0:15:53.280 --> 0:15:55.680 They have to be sort of cautious, and you know, 0:15:55.720 --> 0:15:58.080 I think that's somewhat in there, has become part of 0:15:58.120 --> 0:16:00.240 their DNA. But they've had to get really bold, right 0:16:00.240 --> 0:16:04.080 and they started to release new products. The notebook LM 0:16:04.200 --> 0:16:07.479 is one of the big successes. They're throwing more spaghetti 0:16:07.520 --> 0:16:10.400 against the wall. They're finding that they actually do have courage, right, 0:16:10.440 --> 0:16:13.240 which you know, of course, if you've watched The Wizard 0:16:13.280 --> 0:16:15.600 of Oz, you know in the end they all kind 0:16:15.640 --> 0:16:18.400 of have the thing they thought they were missing. So 0:16:19.000 --> 0:16:23.920 you know, tin Man is Microsoft this well oiled machine. 0:16:24.240 --> 0:16:26.880 If you remember the tin Man was always putting oil 0:16:26.960 --> 0:16:31.680 and oil can. They seem a bit rigid, right, but 0:16:31.720 --> 0:16:33.920 in the end, like they kind of do have a heart. 0:16:33.960 --> 0:16:36.520 I mean I think they I had an interview with 0:16:36.920 --> 0:16:40.400 Satia Nadella recently and and Kevin Scott, the CTO, that 0:16:40.760 --> 0:16:43.120 it's really clear they really do want to sort of 0:16:43.120 --> 0:16:46.000 like enable these developers. I think they have sort of 0:16:46.040 --> 0:16:49.120 a I think a less ego, a bit more of 0:16:49.160 --> 0:16:52.720 a selfless kind of attitude around this stuff. They're building 0:16:52.720 --> 0:16:55.640 the infrastructure for this industry and letting others kind of 0:16:55.920 --> 0:17:00.600 use their creativity to benefit, so they really do have 0:17:00.640 --> 0:17:03.960 a heart in the end. And then I think Amazon, 0:17:04.840 --> 0:17:09.520 you know, this the scarecrow, right, no, no brain. When 0:17:09.560 --> 0:17:11.639 I think about the beginning of this race, you know, 0:17:11.720 --> 0:17:15.000 much like Google, people were really putting Amazon down and 0:17:15.080 --> 0:17:17.960 they thought, oh my god, they've completely missed this, like 0:17:18.040 --> 0:17:22.439 Amazon's got nothing here, and totally ignored the fact that 0:17:22.560 --> 0:17:26.639 actually they had been thinking about this for years. And 0:17:26.720 --> 0:17:31.280 even though you know, maybe they're sort of fighting for 0:17:31.400 --> 0:17:34.520 market share in this new AI race, I think they're 0:17:34.680 --> 0:17:38.679 pretty well positioned to profit from the growth and you know, 0:17:38.800 --> 0:17:42.560 just the the market demand for tokens for compute. 0:17:42.640 --> 0:17:45.320 Are they the market leaders in data centers Amazon today? 0:17:45.560 --> 0:17:48.240 Oh yeah, I mean for sure in the in the cloud, 0:17:48.359 --> 0:17:51.119 in the cloud business. And I think that the question 0:17:51.280 --> 0:17:54.800 though is, you know, can they remain the market leader 0:17:54.880 --> 0:17:57.920 in the AI you know era right, they've been the 0:17:57.960 --> 0:18:00.359 market leader in cloud in terms of just you know, 0:18:00.480 --> 0:18:05.040 normal compute AI data centers is it's a whole new ballgame. 0:18:05.320 --> 0:18:08.360 AI runs on what we would call a GPU, right, 0:18:08.480 --> 0:18:13.040 and so everyone is now you know, racing to add 0:18:13.080 --> 0:18:16.120 these GPUs or build new data centers that are built 0:18:16.119 --> 0:18:19.320 around these GPUs and it's a totally different business model 0:18:19.320 --> 0:18:23.399 and concept, so it kind of restarts the race in 0:18:23.720 --> 0:18:24.240 some ways. 0:18:24.800 --> 0:18:27.679 Okay, so we got Open Air as Dorothy, Microsoft as 0:18:27.720 --> 0:18:31.760 the tin Man, Google's the cowardly line, Amazon is the scarecrow. 0:18:32.119 --> 0:18:35.399 There are also some wizards and witches who need to 0:18:35.440 --> 0:18:38.720 be accounted for Glinda, the Wicked Witch of the West, 0:18:38.760 --> 0:18:41.280 and of course my namesake, the Wizard of Oz. Who's 0:18:41.320 --> 0:18:43.400 who in this cost. 0:18:43.440 --> 0:18:47.439 So let's start with Perplexity being, you know, the wicked Witch. 0:18:48.000 --> 0:18:51.439 As we found in Wicked. You know this, this witch 0:18:51.520 --> 0:18:54.040 was kind of misunderstood, and I think that's sort of 0:18:54.080 --> 0:18:58.080 how Perplexity became the absolute villain, especially in the media industry, 0:18:58.160 --> 0:19:03.120 because they, you know, were were I guess, republishing almost 0:19:03.160 --> 0:19:07.720 like articles without proper citation. I think they very quickly 0:19:08.320 --> 0:19:10.199 changed I mean, it wasn't a policy. I think it 0:19:10.240 --> 0:19:12.159 was a mistake and they very quickly changed them. And 0:19:12.160 --> 0:19:14.880 they're actually have all these media deals where there are 0:19:14.920 --> 0:19:19.360 revenue sharing agreements between Perplexity and the media companies. And 0:19:20.119 --> 0:19:24.080 I think that Wicked image is starting to fade a 0:19:24.160 --> 0:19:27.480 little bit from Perplexity. So I put them in the 0:19:27.560 --> 0:19:32.280 role of the Wicked Witch, and Arvin strinovas the CEO Perplexity, 0:19:32.359 --> 0:19:36.760 responded to me on Twitter with a I think AI 0:19:37.200 --> 0:19:39.919 generated repurposing of the song from Wicked. 0:19:40.160 --> 0:19:43.840 I've got the final close my eyes and leap. It's 0:19:43.880 --> 0:19:46.480 time to try to find Google. I think I'll try 0:19:46.520 --> 0:19:48.960 to find Google and you comple me down. 0:19:50.160 --> 0:19:55.480 Right, So they're disrupting Google. Therefore they're they're really the 0:19:55.600 --> 0:19:57.840 good guys. So you know that was I thought that 0:19:57.920 --> 0:20:01.080 was fun. That was a fun response from Arvind. I think, 0:20:01.280 --> 0:20:03.640 you know, we put Apple as the role of Glinda, 0:20:03.720 --> 0:20:06.440 the good Witch, who you know sort of over over 0:20:06.560 --> 0:20:11.240 promises and underdelivers. I think that's sort of the you know, 0:20:11.560 --> 0:20:13.520 using a little bit of the old and the new 0:20:13.760 --> 0:20:18.960 Wicked version of that that character. Obviously, you know, that's 0:20:19.000 --> 0:20:21.440 based on the fact that they, you know, they had 0:20:21.440 --> 0:20:25.359 this huge event where they promised the world around you know, AI, 0:20:25.480 --> 0:20:27.600 they were going to do these amazing things with Siri 0:20:27.680 --> 0:20:29.800 and your phone, and it turns out, oh, they hadn't 0:20:29.800 --> 0:20:32.000 actually really built any of this, and it's not clear 0:20:32.040 --> 0:20:32.520 that they can. 0:20:33.040 --> 0:20:35.199 And what what's your reporting suggests? Why has it been 0:20:35.240 --> 0:20:35.919 such a bust? 0:20:36.400 --> 0:20:39.560 I think that was a train wreck that anybody who's 0:20:39.560 --> 0:20:41.560 been following Apple has kind of seen for a long 0:20:41.600 --> 0:20:45.360 time because they just have ignored AI and you don't 0:20:45.400 --> 0:20:48.359 see the Apple people are not really like mixing with 0:20:48.400 --> 0:20:50.679 the rest of Silicon Valley, and I think they just 0:20:50.720 --> 0:20:54.880 didn't quite understand this whole AI thing. And and I've 0:20:54.920 --> 0:20:56.919 also written that, you know, I this is a bit 0:20:57.080 --> 0:20:59.159 this is kind of my view, and I don't know, 0:20:59.240 --> 0:21:01.320 I haven't heard that any people agree with me on this, 0:21:01.440 --> 0:21:04.840 but I think they also, you know, because they position 0:21:05.000 --> 0:21:09.800 themselves as this privacy focused company, they that they also 0:21:09.840 --> 0:21:13.399 shied away from AI there because it does require gathering 0:21:13.480 --> 0:21:17.560 a lot of data right on people and personalizing things 0:21:17.560 --> 0:21:19.800 in ways that make people maybe feel like, oh, I'm 0:21:19.840 --> 0:21:23.200 being tracked or what have you. And it turns out 0:21:23.240 --> 0:21:27.000 that that's you know, that's now everything in the tech industry, 0:21:27.080 --> 0:21:28.919 and they're they're way behind. 0:21:29.560 --> 0:21:32.359 And then of course the Wizard, just by having a 0:21:32.400 --> 0:21:35.760 movie named after him, is actually no very important character. 0:21:36.600 --> 0:21:41.679 Right exactly. So I cast x Ai in the in 0:21:41.680 --> 0:21:45.159 the role of the Wizard. You know, I think x 0:21:45.200 --> 0:21:50.040 Ai has this huge, you know, colossus data center, right, 0:21:50.119 --> 0:21:52.760 and they've they've promised that, you know, it really kind 0:21:52.760 --> 0:21:54.760 of made me think of them as that is the 0:21:54.840 --> 0:21:57.000 fact that they're going to understand the secrets of the 0:21:57.080 --> 0:21:59.840 universe with AI that theirs is the Is that literally? 0:22:01.160 --> 0:22:01.400 Yeah? 0:22:01.400 --> 0:22:03.520 I mean this is like literally, you know. I mean 0:22:03.560 --> 0:22:05.800 I saw Elon Musk in a recent interview saying that 0:22:06.240 --> 0:22:10.600 their model grock is based on basic physics and that's 0:22:10.640 --> 0:22:13.480 where it gets its ground truth from. And so you know, 0:22:13.560 --> 0:22:16.840 their effort is to actually ultimately figure out, like, you know, 0:22:16.880 --> 0:22:21.320 a unifying theory of the universe. You know, I think 0:22:21.359 --> 0:22:23.640 that's a lot of it. It remains to be seen 0:22:23.680 --> 0:22:25.000 what's behind the curtain. There. 0:22:25.240 --> 0:22:27.800 Two other companies we haven't spoken about yet are Meta 0:22:28.040 --> 0:22:29.720 and Anthropy. Where do they fit in? 0:22:30.200 --> 0:22:34.840 Yeah, well so Anthropic I cast as in the role 0:22:34.880 --> 0:22:38.160 of the Munchkins. I felt a little bad doing that 0:22:38.240 --> 0:22:40.960