WEBVTT - Tech News: OpenAI's New GPT Won't Destroy The World. Probably. 0:00:04.360 --> 0:00:12.280 Welcome to tech Stuff, a production from iHeartRadio. Be there 0:00:12.280 --> 0:00:15.960 and welcome to tech Stuff. I'm your host Jonathan Strickland, 0:00:15.960 --> 0:00:19.680 Diamond Executive producer with iHeartRadio and how the tech are you. 0:00:20.320 --> 0:00:24.480 It's time for the tech news for Thursday March sixteenth, 0:00:24.960 --> 0:00:29.120 twenty twenty three, and we're back on the Big Topic 0:00:29.720 --> 0:00:34.680 of twenty twenty three AI, although artificial intelligence should really 0:00:34.680 --> 0:00:38.120 watch it's back because if more tech related financial institutions 0:00:38.120 --> 0:00:40.320 fail this year, I think there's going to be a 0:00:40.320 --> 0:00:43.200 new contender for the Big Topic of twenty twenty three. 0:00:43.720 --> 0:00:49.120 But this week open ai announced the release of GPT four. 0:00:49.640 --> 0:00:54.120 This is the large language model that is now multimodal. 0:00:54.600 --> 0:00:56.480 All right, let's break this down. So first of all, 0:00:56.560 --> 0:01:02.600 GPT stands for generative pre train transformer. The word transformer 0:01:02.600 --> 0:01:06.000 in this case doesn't refer to either a robot in 0:01:06.040 --> 0:01:09.600 disguise or the device that can step voltage up or 0:01:09.720 --> 0:01:13.560 down with alternating current. Instead, this kind of transformer is 0:01:13.600 --> 0:01:17.720 a type of machine learning model that Google first introduced 0:01:17.760 --> 0:01:22.560 back in twenty seventeen. The word generative indicates that it's 0:01:22.600 --> 0:01:27.360 capable of creating something generating something as for a multimodal 0:01:27.959 --> 0:01:31.760 that means this new version of GPT can accept different 0:01:31.840 --> 0:01:37.679 kinds of input prompts to create output. So previously GPT 0:01:37.760 --> 0:01:42.360 could only accept text, but now it can also accept images, 0:01:42.480 --> 0:01:47.200 and it can give analysis of those images or presumably 0:01:47.319 --> 0:01:51.280 incorporate those images into responses, like maybe you could ask 0:01:51.400 --> 0:01:57.960 GPT who is the person in this picture? Or how 0:01:58.000 --> 0:02:01.120 many elements are in this photograph? Or can you tell 0:02:01.160 --> 0:02:03.360 me a story about what is in this picture? That 0:02:03.440 --> 0:02:08.720 kind of stuff. GPT four also seriously ups the word 0:02:08.800 --> 0:02:14.400 count on its responses. So the previous version GPT three 0:02:14.480 --> 0:02:18.200 point five, that's the version that actually powers chat GPT, 0:02:19.000 --> 0:02:23.040 it could create a response of around three thousand words max. 0:02:23.520 --> 0:02:27.799 But GPT four can get positively low quacious and provide 0:02:28.040 --> 0:02:31.160 deep responses to queries that could go up to twenty 0:02:31.240 --> 0:02:35.040 five thousand words. So now we're getting into like Jonathan 0:02:35.080 --> 0:02:38.920 in a casual conversation level here text stuff episode. In 0:02:38.919 --> 0:02:43.480 other words, it's also apparently better at playing by the 0:02:43.560 --> 0:02:46.440 rules that open ai has set. That is, it is 0:02:46.560 --> 0:02:51.120 less likely than chat GPT three point five to respond 0:02:51.120 --> 0:02:56.200 to requests that would break policy, you know, stuff like 0:02:56.880 --> 0:03:02.840 generating hate speech or responding to requests to cause problems 0:03:02.840 --> 0:03:06.440 within a community, it is less likely to do that, 0:03:06.720 --> 0:03:12.040 not completely impervious to it, but it's better than the 0:03:12.120 --> 0:03:15.639 previous generations were. Developers will also have a lot more 0:03:15.639 --> 0:03:18.080 options when they use GPT as part of their apps. 0:03:18.880 --> 0:03:22.000 They can even shape how the AI will respond to 0:03:22.040 --> 0:03:24.480 a user. They can adjust stuff like the tone and 0:03:24.560 --> 0:03:29.079 the style of responses. So if you have something that's 0:03:29.120 --> 0:03:31.720 like a say, a fun and silly app that for 0:03:31.760 --> 0:03:34.079 some reason needs to tap into the power of GPT, 0:03:34.960 --> 0:03:39.080 you could adjust the tone to be more lighthearted and 0:03:39.160 --> 0:03:42.520 playful rather than stick with the standard tone and style 0:03:43.200 --> 0:03:47.080 of the normal response, which is what previous generations of 0:03:47.160 --> 0:03:51.280 GPT were stuck with. And in a demonstration, open Ai 0:03:51.360 --> 0:03:55.960 showed that GPT four can even do some pretty astonishing stuff, 0:03:56.000 --> 0:04:01.080 some advanced tasks. So, for example, you could write an 0:04:01.240 --> 0:04:04.320 idea for a basic website right like you're using a 0:04:04.360 --> 0:04:07.920 notepad and you're writing out your concept or your website. 0:04:08.280 --> 0:04:11.080 You can then feed that concept to GPT and it 0:04:11.120 --> 0:04:15.640 could actually create the website complete with basic functionality. So 0:04:15.680 --> 0:04:19.719 the example I saw had GPT produce a website based 0:04:19.760 --> 0:04:22.560 around jokes, so you would get the setup of the 0:04:22.640 --> 0:04:26.640 joke appearing as text and then to get the punchline, 0:04:26.640 --> 0:04:28.760 you would have to click on a little box that 0:04:28.800 --> 0:04:33.359 would reveal what the punchline was to the joke. And 0:04:33.440 --> 0:04:37.040 this was all prompted by some pretty simple notes about 0:04:37.120 --> 0:04:42.279 the parameters for this website. Like it wasn't copying the note, 0:04:42.279 --> 0:04:44.599 it was taking the notes as directions and then making 0:04:44.640 --> 0:04:48.080 the website based on those directions. That's pretty cool. Now, 0:04:48.080 --> 0:04:50.520 it still has a lot of the problems of earlier 0:04:50.560 --> 0:04:55.279 generations of GPT. Its responses can still be inaccurate and misleading. 0:04:56.360 --> 0:05:00.479 The common term I've seen in this are hallucinations, where 0:05:00.680 --> 0:05:04.640 the response ends up not being an accurate representation. It's 0:05:04.680 --> 0:05:09.440 a hallucination. I don't really like that term personally. I 0:05:09.560 --> 0:05:13.920 just I think it ends up kind of sugarcoating what 0:05:13.960 --> 0:05:17.880 we're actually talking about here, which is trustworthiness and accuracy 0:05:18.400 --> 0:05:22.640 and hallucination. I don't know, it seems a little too 0:05:23.120 --> 0:05:26.040 it seems like it's letting GPT off the hook a 0:05:26.080 --> 0:05:29.840 little too much. That's my own personal opinion. It's kind 0:05:29.839 --> 0:05:32.240 of similar to how I feel about Tesla and naming 0:05:32.360 --> 0:05:35.400 its driver assist feature full self driving. I don't think 0:05:35.440 --> 0:05:38.200 that is an accurate representation of what it actually does 0:05:38.760 --> 0:05:42.080 same sort of thing, right, Like names and words mean things, 0:05:42.760 --> 0:05:45.680 and sometimes you have to ask why are these words 0:05:45.720 --> 0:05:49.400 being used to describe this? Is it in an attempt 0:05:49.839 --> 0:05:53.600 to kind of take some pressure off. And that's just 0:05:53.640 --> 0:05:57.240 my own personal take on this. I could be way 0:05:57.240 --> 0:06:03.160 way off base. Anyway, it's still pretty cool. But obviously, 0:06:03.200 --> 0:06:08.240 if GPT produces information that is misleading or inaccurate, but 0:06:08.400 --> 0:06:12.200 the presentation is put across in such a way as 0:06:12.240 --> 0:06:16.000 to seem authentic, that's a real problem. And it could 0:06:16.000 --> 0:06:19.039 also still be used to do stuff like promote ideologies 0:06:19.040 --> 0:06:23.880 and create propaganda in ways that are perhaps not entirely ethical. 0:06:24.040 --> 0:06:27.760 So while the new features are really impressive, the challenges 0:06:27.800 --> 0:06:30.320 that we face with these kinds of AI models haven't 0:06:30.880 --> 0:06:35.600 magically been dismissed. They still very much exist. However, one 0:06:35.680 --> 0:06:38.560 issue that apparently we do not need to worry about 0:06:38.839 --> 0:06:43.040 is GPT for's ability to destroy us all. So I'm 0:06:43.080 --> 0:06:46.520 being a bit glib here, but Open Ai actually did 0:06:46.520 --> 0:06:51.160 allow an AI research group called the Alignment Research Center 0:06:51.360 --> 0:06:55.160 or ARC, to test the Large Language Model to make 0:06:55.200 --> 0:06:58.680 sure it doesn't show signs of being capable of starting 0:06:58.680 --> 0:07:01.400 the robot uprising and learning all of us into batteries 0:07:01.480 --> 0:07:05.840 or something Specifically, the AI group was looking into possibilities 0:07:05.839 --> 0:07:12.360 like the model's capacity to copy itself or to acquire resources, 0:07:12.400 --> 0:07:16.720 whether digital or real world or whatever, or maybe its 0:07:16.720 --> 0:07:20.120 ability to manipulate people with stuff like phishing attacks and 0:07:20.200 --> 0:07:23.200 that kind of thing. Also whether or not it can 0:07:23.280 --> 0:07:26.440 make high level plans and then maybe follow through on 0:07:26.480 --> 0:07:29.040 those plans. So this kind of gets into some of 0:07:29.040 --> 0:07:32.080 the scary stuff around AI that we typically think of 0:07:32.120 --> 0:07:35.920 as being science fiction, right, like Terminator style stuff. The 0:07:36.000 --> 0:07:40.040 thought of a sufficiently intelligent AI could be capable of 0:07:40.200 --> 0:07:43.040 escaping whatever constraints we try to put on it, because 0:07:43.080 --> 0:07:46.960 by definition, if it's super intelligent, then it's smarter than 0:07:47.000 --> 0:07:50.280 we are, right, So if it's smarter than we are, 0:07:50.400 --> 0:07:52.480 then there it stands to reason it would figure out 0:07:52.520 --> 0:07:54.880 a way to get out of whatever box we try 0:07:54.880 --> 0:07:58.760 to put it in, including the possibility of manipulating people 0:07:58.880 --> 0:08:02.400 to essentially set it free. There have been thought experiments 0:08:02.440 --> 0:08:07.920 that have shown this is entirely possible. The AI researchers 0:08:07.960 --> 0:08:12.280 concluded that GPT four doesn't measure up to this kind 0:08:12.320 --> 0:08:16.080 of dangerous AI, But then, as Ours Technica reports, there 0:08:16.120 --> 0:08:19.120 are scarce details on the nature of the tests that 0:08:19.200 --> 0:08:23.040 the research group actually conducted, so it's hard to judge 0:08:23.040 --> 0:08:28.000 whether or not their findings are trustworthy themselves. Also, I 0:08:28.040 --> 0:08:30.360 think we need to remind ourselves that whether or not 0:08:30.440 --> 0:08:35.479 AI poses an existential threat to humanity, we can acknowledge 0:08:35.480 --> 0:08:39.640 that AI contributes to real problems in the here and now. Now. 0:08:39.760 --> 0:08:43.440 Usually this comes as an unintended consequence, such as the 0:08:43.440 --> 0:08:46.720 inclusion of bias within a system that leads to one 0:08:46.720 --> 0:08:51.760 group of people being disproportionately affected by negative outcomes, facial 0:08:51.800 --> 0:08:55.400 recognition technologies being a great example of that, but it 0:08:55.440 --> 0:08:58.680 could also be included by design, right, someone could make 0:08:58.800 --> 0:09:03.520 an AI applicant designed to create harm in some way. 0:09:04.000 --> 0:09:07.360 The alignment part of Alignment Research Center refers to the 0:09:07.440 --> 0:09:10.920 goal of making sure that AI is in alignment or 0:09:11.000 --> 0:09:14.880 agreement with basic principles that are meant to benefit humanity. 0:09:15.400 --> 0:09:18.080 But here's the thing. Bias can come into that sort 0:09:18.080 --> 0:09:20.360 of decision making process too, and you have to ask 0:09:20.400 --> 0:09:24.160 tough questions of who decides that criteria, who is it 0:09:24.200 --> 0:09:28.800 that's defining what is beneficial because believe me, there is 0:09:28.840 --> 0:09:32.560 not a universal or objective truth to what that actually 0:09:32.880 --> 0:09:36.880 even means. And even an effort to create an quote 0:09:36.920 --> 0:09:41.120 unquote aligned AI could result in one that benefits a 0:09:41.160 --> 0:09:46.080 small group at the expense of everyone else. Earlier this year, 0:09:46.120 --> 0:09:49.600 I mentioned that the Chinese megacompany by Do was hard 0:09:49.600 --> 0:09:52.839 at work on its own aipowered chatbot that it had 0:09:52.880 --> 0:09:56.280 called ernie Bot. Well earlier Today, by Do showed off 0:09:56.440 --> 0:09:59.840 ernie Bot, but did so with prerecorded videos in showing 0:10:00.000 --> 0:10:03.720 off its capabilities. Obviously, prerecorded videos can be fudged a 0:10:03.720 --> 0:10:07.280 little bit, so that brought some concern. It also didn't 0:10:07.320 --> 0:10:10.200 launch any sort of public access to the tool. It's 0:10:10.240 --> 0:10:14.760 keeping access very much limited. And dissatisfaction followed, and Bydu's 0:10:14.880 --> 0:10:18.280 stock shares dropped as much as ten percent initially, though 0:10:18.320 --> 0:10:20.959 that leveled off a little bit later, so the company 0:10:21.120 --> 0:10:26.720 only lost around three billion dollars in value. Yikes. Bydu 0:10:26.800 --> 0:10:30.120 seems to be taking a more measured approach to deploying 0:10:30.120 --> 0:10:34.480 its AI chatbot, which I actually argue is the responsible 0:10:34.520 --> 0:10:37.120 thing to do. We have seen numerous stories of how 0:10:37.200 --> 0:10:43.479 GPT has caused problems with premature deployments, such as homophobic 0:10:43.559 --> 0:10:47.600 jokes on the endless computer generated Seinfeld episode that's on Twitch. 0:10:48.200 --> 0:10:51.760 All way to creating a concern among teachers that their 0:10:51.760 --> 0:10:54.760 students could be using a chatbot to cheat on assignments, 0:10:54.800 --> 0:10:58.360 and more so, a more deliberate rollout seems like it's 0:10:58.360 --> 0:11:01.079 a good idea to me. But then I guess there's 0:11:01.080 --> 0:11:04.600 this perception that Baidu is trailing behind companies like Microsoft 0:11:04.679 --> 0:11:07.960 and Google, and there's this fear that it's not being competitive. 0:11:08.600 --> 0:11:12.000 I think not being competitive in this case also means 0:11:12.000 --> 0:11:16.760 you're sidestepping some potentially really troubling problems down the line. 0:11:17.240 --> 0:11:21.720 But I'm not an investor, So there you go. Okay, 0:11:22.040 --> 0:11:24.480 we've got a lot more stories to cover, but before 0:11:24.480 --> 0:11:37.120 we get to that, let's take a quick break. We're back, 0:11:37.480 --> 0:11:40.640 so let's talk about TikTok. We've got a couple of 0:11:40.679 --> 0:11:43.880 stories about that. Again. The government of the United Kingdom 0:11:44.080 --> 0:11:48.400 is following the latest trend of banning TikTok on government 0:11:48.440 --> 0:11:51.839 owned devices, or at least it's considering that option. By 0:11:51.880 --> 0:11:54.560 the time you hear this podcast, the matter may have 0:11:55.000 --> 0:11:57.920 officially been decided, but as I was recording it, it 0:11:58.000 --> 0:12:01.440 was still something that was being considered, but we were 0:12:01.520 --> 0:12:05.000 expecting an announcement sometime today. The UK would follow in 0:12:05.040 --> 0:12:09.000 the footsteps of the United States, Canada, the Netherlands, the 0:12:09.040 --> 0:12:12.520 European Union in general, and several other countries if they 0:12:12.559 --> 0:12:15.280 were to do this, and yes, the heart of the 0:12:15.320 --> 0:12:19.199 matter is the security concerns of data potentially filtering to 0:12:19.360 --> 0:12:24.040 TikTok's Chinese owned parent company, Byte Dance, and by extension, 0:12:24.360 --> 0:12:28.320 to the Chinese government. Yesterday, The Wall Street Journal reported 0:12:28.559 --> 0:12:32.559 that the United States could send TikTok and byte Dance 0:12:32.559 --> 0:12:36.679 and ultimatum, which is essentially, you must sever all ties 0:12:36.720 --> 0:12:39.360 between TikTok and bitte Dance. By Dance would have to 0:12:40.120 --> 0:12:43.280 spin off TikTok as an entirely independent company and not 0:12:43.480 --> 0:12:49.600 be the parent company anymore, or TikTok risks being banned nationwide. 0:12:49.960 --> 0:12:52.960 This echoes what former President Donald Trump tried to do 0:12:53.040 --> 0:12:56.080 a few years ago. He tried to force TikTok to 0:12:56.200 --> 0:12:59.280 divorce itself from its parent company and become a truly 0:12:59.440 --> 0:13:04.360 US organization. During Trump's presidency, the push was for TikTok 0:13:04.400 --> 0:13:08.160 to sell itself to some larger American company, but this 0:13:08.280 --> 0:13:11.600 never came to fruition, partly because TikTok argued in court 0:13:12.120 --> 0:13:15.200 that this directive from the president would violate a law 0:13:15.280 --> 0:13:20.800 called the Burman amendments that prevent presidents from using economic 0:13:20.880 --> 0:13:29.640 pressures to force international communications platforms from doing stuff. So essentially, 0:13:29.640 --> 0:13:32.040 they were saying this is against the law, and now 0:13:32.080 --> 0:13:34.719 we're seeing a similar push from the Biden administration. I'm 0:13:34.760 --> 0:13:38.240 not exactly sure how that same law wouldn't apply in 0:13:38.280 --> 0:13:40.800 this case, Like, I don't know that there's a new 0:13:40.960 --> 0:13:45.520 argument to be made if it's coming straight from the president. Anyway, 0:13:45.640 --> 0:13:49.040 TikTok reps argue that a massive one and a half 0:13:49.160 --> 0:13:52.720 billion dollar program meant to secure data in the US 0:13:53.160 --> 0:13:55.920 and to allow third party security companies a chance to 0:13:56.080 --> 0:13:59.840 monitor TikTok's operations to look for anything hinky going on. 0:14:00.640 --> 0:14:03.760 They're saying that's more than enough to solve the worries 0:14:03.760 --> 0:14:06.360 that TikTok is serving as a threat to national security. 0:14:06.800 --> 0:14:10.320 And then further, reps for TikTok have said that if 0:14:10.320 --> 0:14:13.760 by Dance did divest itself of TikTok, that wouldn't actually 0:14:13.800 --> 0:14:17.199 prevent the transmission of data from TikTok to China. I'm 0:14:17.240 --> 0:14:20.080 not entirely sure how all this tracks, but then I 0:14:20.160 --> 0:14:23.560 haven't been given a glimpse of all the details, so 0:14:23.680 --> 0:14:25.680 I doubt I would even understand all of it if 0:14:25.680 --> 0:14:29.480 I did, because this is wrapping into things like finance 0:14:29.560 --> 0:14:33.240 and trade more than the technology itself. Anyway, it looks 0:14:33.280 --> 0:14:36.560 like we're heading to an impass and I'm not entirely 0:14:36.640 --> 0:14:39.720 sure where it's going to go from there. But my 0:14:39.840 --> 0:14:42.280 guess is we're not going to see the pressure on 0:14:42.320 --> 0:14:48.960 TikTok going away anytime soon. Hey, y'all remember FTX, You know, 0:14:49.240 --> 0:14:51.640 the company that used to be the second largest crypto 0:14:51.720 --> 0:14:54.680 exchange in the world before word got out that the 0:14:54.720 --> 0:14:58.960 team behind it was practicing some creative accounting that you know, 0:14:59.120 --> 0:15:01.360 you might be able to look at another point of 0:15:01.400 --> 0:15:05.280 view and call it, you know, fraud. But more bad 0:15:05.280 --> 0:15:08.480 news has emerged about the company and its high profile 0:15:08.560 --> 0:15:13.680 co founder, Sam Bankman Freed aka SBF. So the new 0:15:13.720 --> 0:15:16.680 folks in charge of FTX, who are mostly going through 0:15:16.720 --> 0:15:19.320 the process of selling the company off for parts in 0:15:19.440 --> 0:15:22.840 order to return as much value to investors and customers 0:15:22.840 --> 0:15:26.480 as possible, have said that they have uncovered transfers that 0:15:26.600 --> 0:15:31.640 collectively amounted to around two point two billion dollars and 0:15:31.680 --> 0:15:35.680 that this money was sent to SBF through various means, 0:15:35.680 --> 0:15:40.200 primarily through Alameda Research, the hedge fund company that was 0:15:42.000 --> 0:15:48.160 you know Yen to ftx's yang. An additional billion seemed 0:15:48.280 --> 0:15:50.720 to be pulled from FTX to go to other key 0:15:50.800 --> 0:15:57.720 members of FTX, and yeah, Alameda Research handled those transfers, 0:15:57.760 --> 0:16:02.160 according to the new owners of T and that you know, 0:16:02.600 --> 0:16:06.320 it's been going on for a while. And I personally 0:16:06.320 --> 0:16:10.400 consider the report shocking but not surprising. By that, I mean, 0:16:10.520 --> 0:16:13.040 I'm not at all surprised that SPF was trying to 0:16:13.080 --> 0:16:15.720 pull as much cash out as he could before FTX 0:16:15.880 --> 0:16:19.680 came crashing down. It is shocking exactly how much money 0:16:19.760 --> 0:16:23.920 that was, right like to pull out two point two 0:16:24.040 --> 0:16:28.840 billion dollars in various means as you're trying to salvage 0:16:28.840 --> 0:16:30.880 as much as you can from a ship that's sinking. 0:16:31.560 --> 0:16:36.200 Incredible Anyway. SPF is currently out on bail awaiting his trial, which, 0:16:36.240 --> 0:16:39.920 according to the current court schedule, will not begin until 0:16:39.960 --> 0:16:47.080 October this year because justice is super swift. The Verge 0:16:47.080 --> 0:16:51.640 has an article titled the Silicon Valley bank fallout is 0:16:51.720 --> 0:16:54.240 just beginning. It's a great article. You should go read 0:16:54.280 --> 0:16:57.080 it on the Verge and the articles mostly about how 0:16:57.120 --> 0:16:59.720 tech companies are trying to figure out where to go 0:16:59.760 --> 0:17:03.360 for word from here, what's the safest place to bank 0:17:03.920 --> 0:17:06.800 now that SVB has gone under, because if you remember, 0:17:06.960 --> 0:17:11.000 SVB was like a financial pillar for the tech industry. 0:17:11.680 --> 0:17:15.040 So this piece lays out the challenges that tech companies 0:17:15.080 --> 0:17:19.199 currently face, including how to mitigate risk while ensuring that 0:17:19.280 --> 0:17:23.280 tasks like making payroll aren't interrupted. Like you might argue, hey, 0:17:23.760 --> 0:17:26.960 it'd be really good to make sure you're depositing in 0:17:27.040 --> 0:17:30.199 different banks so that way you're not having all of 0:17:30.200 --> 0:17:32.720 your eggs in one basket, and if that basket goes 0:17:32.800 --> 0:17:36.119 belly up, you're still all right because you've got eggs 0:17:36.119 --> 0:17:38.639 and other baskets, right, Except that if you're trying to 0:17:38.640 --> 0:17:42.680 do things like pay out a large group of employees, 0:17:43.200 --> 0:17:46.600 having that money dispersed makes it much more difficult to 0:17:46.640 --> 0:17:49.520 do those kind of tasks. So that's the sort of 0:17:49.600 --> 0:17:54.600 challenges that tech companies are looking at. The piece also 0:17:54.680 --> 0:17:58.119 mentions that we're probably going to see a change in 0:17:58.280 --> 0:18:02.760 how startups are funded and how they operate, and part 0:18:02.760 --> 0:18:05.720 of me thinks that a reevaluation of startup culture is 0:18:06.040 --> 0:18:10.359 long overdue. I have often worried that the tendency for 0:18:10.400 --> 0:18:14.200 startups to launch without first creating a firm business plan, 0:18:15.200 --> 0:18:18.400 instead essentially having a desire to get acquired by some 0:18:18.480 --> 0:18:22.480 other company has really led to irresponsible behavior, and that 0:18:22.600 --> 0:18:26.240 this has gone unchecked for too long. This created too 0:18:26.240 --> 0:18:31.199 many companies that ultimately produced very little value. Right. You'll 0:18:31.240 --> 0:18:35.200 hear about a company like Google or Meta or Amazon 0:18:35.560 --> 0:18:39.640 scooping up some of these companies, and sometimes nothing really 0:18:39.640 --> 0:18:42.320 seems to come of it, right, and you're just like, well, 0:18:42.720 --> 0:18:45.879 what good did that startup due apart from make the 0:18:46.240 --> 0:18:50.720 founders rich? And then you see the founders go off 0:18:50.720 --> 0:18:52.679 and do the same thing again, where they'll start up 0:18:52.680 --> 0:18:57.200 a new company, perhaps one that also has very little 0:18:57.440 --> 0:19:01.199 business plan element to it, also with the hope that 0:19:01.240 --> 0:19:03.040 they are going to get scooped up by another company, 0:19:03.200 --> 0:19:05.480 and they do it again and again and again, because 0:19:05.560 --> 0:19:08.520 coming up with an idea that sounds attractive is way 0:19:08.560 --> 0:19:11.840 easier than making that idea work. And if you are 0:19:11.880 --> 0:19:16.119 good at selling ideas, you can just do a serial 0:19:16.160 --> 0:19:19.240 approach to create cool sounding idea, sell it off to 0:19:19.280 --> 0:19:24.840 someone else, rents, and repeat and by an island somewhere, 0:19:25.520 --> 0:19:30.280 I'm hoping that this represents a reality check moment where 0:19:30.280 --> 0:19:32.560 we will see a more thoughtful and careful approach to 0:19:32.680 --> 0:19:38.840 funding startups that might be a vain hope, because you know, 0:19:39.320 --> 0:19:43.720 historically investors have shown a really strong desire to get 0:19:43.760 --> 0:19:47.000 in on the thing that's possibly going to take the 0:19:47.000 --> 0:19:50.040 world by storm, so that you get so rich that 0:19:50.119 --> 0:19:53.960 you make Scrooge mcdock look like a pauper. Anyway, the 0:19:54.119 --> 0:19:55.960 article in The Verge is a good one again, it's 0:19:55.960 --> 0:19:59.280 called the Silicon Valley Bank fallout is just beginning. It 0:19:59.320 --> 0:20:01.639 also takes time to point out that the collapse of 0:20:01.840 --> 0:20:05.359 SVB was in large part the fault of some of 0:20:05.400 --> 0:20:10.800 the venture capitalist investors themselves. Like you had the leaders 0:20:10.880 --> 0:20:14.680 of big investment fund groups saying, hey, you should probably 0:20:14.680 --> 0:20:17.040 pull your money out of SVB because if you don't, 0:20:17.080 --> 0:20:18.480 someone else is going to do it, and then you 0:20:18.480 --> 0:20:20.159 won't be able to get your money, and so they 0:20:20.240 --> 0:20:23.119 ended up creating the very crisis that they were warning 0:20:23.160 --> 0:20:27.840 people about. So I think a lot of financial institutions 0:20:27.920 --> 0:20:31.160 may still view loans to the tech sector as being 0:20:31.240 --> 0:20:35.960 risky because the venture capital community have proven themselves to 0:20:36.040 --> 0:20:42.080 be self destructive. It's hard to trust in an industry 0:20:42.160 --> 0:20:48.840 when you see the tendency to act in self interest 0:20:49.119 --> 0:20:53.240 to the point where you harm everybody else. Very difficult 0:20:53.240 --> 0:20:56.919 to put your trust into that group. Okay, still have 0:20:56.960 --> 0:20:59.439 a few more stories to cover before I get to that. 0:20:59.560 --> 0:21:13.560 Let's another quick break, Okay. The Federal Trade Commission or FTC, 0:21:13.880 --> 0:21:17.960 has finalized its judgment against Epic Games regarding the company's 0:21:18.000 --> 0:21:22.840 practice of enticing or perhaps even fooling Fortnite players into 0:21:22.880 --> 0:21:28.240 making in game purchases through what the FTC calls dark patterns. 0:21:28.400 --> 0:21:32.919 So essentially, dark patterns refers to creating an interface that 0:21:33.000 --> 0:21:36.520 makes it really easy, far too easy, most would argue, 0:21:36.680 --> 0:21:39.280 for a player to make an in game purchase, perhaps 0:21:39.280 --> 0:21:43.639 without even knowing or understanding that it is a financial transaction. 0:21:44.280 --> 0:21:46.960 One of the ways the FTC argued Epic Games did 0:21:47.000 --> 0:21:51.040 this was that it ended up playing fast and loose 0:21:51.119 --> 0:21:55.040 with button inputs, so that you might have a button 0:21:55.280 --> 0:21:58.959 that normally has you back out of a menu option, right, Like, 0:21:59.400 --> 0:22:03.199 maybe you've had your menu UI set up so that 0:22:03.240 --> 0:22:07.040 when you hit B, you're backing out of that part 0:22:07.040 --> 0:22:08.680 of the menu and you go to a larger part. 0:22:09.280 --> 0:22:12.399 But then in a transaction, maybe you suddenly make B 0:22:12.840 --> 0:22:16.080 confirm instead of backout, and so people who are used 0:22:16.119 --> 0:22:19.480 to using bing and backout hit B, but now they've 0:22:19.480 --> 0:22:24.280 just made a payment, and that might not be something 0:22:24.320 --> 0:22:27.200 you could easily reverse or cancel out of, like maybe 0:22:27.240 --> 0:22:31.040 there's no confirmed feature. You just boom, you've done it. 0:22:31.560 --> 0:22:33.320 Another part of the problem is that a lot of 0:22:33.359 --> 0:22:35.760 the folks who are playing Fortnite are kids, So the 0:22:35.840 --> 0:22:38.919 FTC argued that Epic Games didn't include enough protections to 0:22:38.960 --> 0:22:42.840 prevent kids from making in game purchases without parental consent, 0:22:43.480 --> 0:22:46.800 so they just started racking up enormous charges on parents 0:22:46.880 --> 0:22:49.720 credit cards that were associated with the account, and if 0:22:49.720 --> 0:22:52.360 a player did go so far as to contact their 0:22:52.359 --> 0:22:56.640 credit card company to dispute charges, Epic Games would then 0:22:56.800 --> 0:22:59.560 lock that person's player account so they couldn't play the 0:22:59.560 --> 0:23:02.800 game anymore. So now the FTC is telling Epic Games 0:23:02.840 --> 0:23:07.000 to cough up two hundred forty five million dollars as punishment. 0:23:07.320 --> 0:23:10.480 The FTC plans to use that money to provide refunds 0:23:10.520 --> 0:23:14.640 to affected players. Plus, Epic will not be allowed to 0:23:14.680 --> 0:23:18.159 block people who dispute credit card charges anymore per the 0:23:18.280 --> 0:23:22.200 terms of this agreement. If you think you were affected 0:23:22.200 --> 0:23:25.480 by Epic Games practices. In other words, if you feel 0:23:25.480 --> 0:23:28.000 you were tricked into making a purchase on Fortnite and 0:23:28.160 --> 0:23:30.560 you are a US citizen, you can go to the 0:23:30.560 --> 0:23:37.240 website FTC dot gov slash fortnite that's foart n Ie, 0:23:37.640 --> 0:23:39.960 and you can fill in a little online form to 0:23:39.960 --> 0:23:44.000 start the refund process. South Korea and Samsung announced plans 0:23:44.000 --> 0:23:49.160 to build a truly enormous semiconductor manufacturing campus this week. 0:23:49.560 --> 0:23:53.119 It's one that Samsung is going to invest two hundred 0:23:53.160 --> 0:23:57.680 thirty billion dollars in two The effort will make South 0:23:57.760 --> 0:24:02.480 Korea much more competitive with Taiwan's TSMC as a semiconductor 0:24:02.520 --> 0:24:06.600 manufacturing company that's responsible for more than half of all 0:24:06.640 --> 0:24:10.800 the semiconductors used in advanced electronics today. It's also a 0:24:10.840 --> 0:24:15.359 move to create a better supply in a world that 0:24:15.440 --> 0:24:18.280 not so long ago was in a serious crunch due 0:24:18.320 --> 0:24:20.320 to many factors, in the big one being the COVID 0:24:20.400 --> 0:24:24.879 nineteen pandemic. Samsung was one of several companies hit hard 0:24:24.920 --> 0:24:28.520 by supply chain issues, and by investing in semiconductor manufacturing 0:24:28.560 --> 0:24:31.600 facilities within South Korea, there are hopes to head off 0:24:31.640 --> 0:24:33.840 future problems like what we saw on the recent past. 0:24:34.320 --> 0:24:36.159 This is not that different from what we're seeing here 0:24:36.160 --> 0:24:39.360 in the United States. The US government has poured billions 0:24:39.400 --> 0:24:42.080 of dollars or at least earmarked billions of dollars for 0:24:42.160 --> 0:24:46.080 similar investments here in the US again to try and 0:24:46.160 --> 0:24:50.359 alleviate some of these bottleneck points for supply chain issues 0:24:50.359 --> 0:24:54.280 in the future. Over in Japan, the US startup Lift 0:24:54.359 --> 0:24:58.760 Aircraft Incorporated held its first test flight of the Hexa 0:24:59.320 --> 0:25:04.399 flying car vehicle with an actual human being piloting it. 0:25:04.760 --> 0:25:07.399 So it was the first time it had a piloted 0:25:07.600 --> 0:25:11.280 test flight. So it looks like like an oversized quad 0:25:11.280 --> 0:25:14.560 copter that has a cockpit in it that can hold 0:25:14.640 --> 0:25:18.040 a single rider. Now, the plan is to have flying 0:25:18.119 --> 0:25:21.080 vehicles give rides to passengers in time for the twenty 0:25:21.119 --> 0:25:25.160 twenty five Osaka Kansai Expo. And from what I read, 0:25:25.920 --> 0:25:28.239 it is a single seater vehicle, so I'm guessing that 0:25:28.280 --> 0:25:32.199 for passenger rides it must be either controlled remotely or 0:25:32.760 --> 0:25:36.480 possibly autonomous. I think remote control is far more realistic, 0:25:37.000 --> 0:25:40.880 but there's no way you would just hand over control 0:25:40.960 --> 0:25:43.359 of a flying car to some rando and then just 0:25:43.440 --> 0:25:47.920 say good luck out there. Anyway, this particular vehicle isn't 0:25:47.960 --> 0:25:51.720 meant for the long haul because according to Asahi dot Com, 0:25:52.240 --> 0:25:55.920 it can fly for about fifteen minutes, so you don't 0:25:55.960 --> 0:25:58.000 even get twenty minutes of travel out of this thing, 0:25:58.080 --> 0:26:01.760 so it's meant for very short distance as it travels 0:26:01.760 --> 0:26:04.760 at speeds top speeds at around one hundred kilometers per 0:26:04.760 --> 0:26:08.680