WEBVTT - Spot the Bot 0:00:00.160 --> 0:00:07.080 Brought to you by Toyota. Let's go places. Welcome to 0:00:07.280 --> 0:00:14.280 Forward Thinking, either you want and Welcome to Forward Thinking, 0:00:14.400 --> 0:00:17.200 the podcast that looks at the future and says, one 0:00:17.239 --> 0:00:20.400 more robot learns to be something more than a machine. 0:00:20.720 --> 0:00:25.320 I'm Jonathan Strickland and I'm Joe McCormick, and I've got 0:00:25.320 --> 0:00:29.480 a question for y'all. Would you like to ask it? 0:00:29.640 --> 0:00:32.520 How how unusual? Please ask your question? Have you ever 0:00:32.640 --> 0:00:36.800 noticed how any time there's a human being who follows 0:00:36.840 --> 0:00:39.720 you on Twitter or favorites one of your tweets, and 0:00:39.800 --> 0:00:42.680 you click through to their profile, and then you click 0:00:42.760 --> 0:00:46.159 on their website, it's just some spam page with a 0:00:46.159 --> 0:00:50.159 bunch of pop up ads. I will say that I 0:00:50.200 --> 0:00:52.840 am not unfamiliar with what you are saying. I try 0:00:52.880 --> 0:00:56.480 not to click through to the website. Yeah, that's a 0:00:56.560 --> 0:01:00.279 smart move, Lauren, because no, I don't notice the as 0:01:00.280 --> 0:01:03.040 pervasively as I was just leading you to believe. But 0:01:03.640 --> 0:01:06.840 there is a certain phenomenon that I think we are 0:01:06.880 --> 0:01:11.680 all somewhat familiar with now, the fact that there are 0:01:12.200 --> 0:01:18.000 let's call them agents loose on the web that are 0:01:18.040 --> 0:01:22.200 not human beings and and not like subhuman beings. We're 0:01:22.200 --> 0:01:25.480 not being elitist in some kind of way. This isn't 0:01:25.520 --> 0:01:30.080 an episode about the future of trolls. Yeah, there there 0:01:30.120 --> 0:01:33.120 are people who lack humanity, but we we would probably 0:01:33.160 --> 0:01:36.720 say that they possess consciousness and like a brain and 0:01:36.880 --> 0:01:40.080 probably some kind of flesh body. Yeah. Yeah, there are 0:01:40.120 --> 0:01:43.960 many agents indeed on the Internet that are not possessed 0:01:44.160 --> 0:01:48.240 of a flesh body. Yeah. This isn't a new thing either, No, no, no, no. 0:01:48.320 --> 0:01:50.840 And so we've talked plenty of times on this podcast 0:01:50.880 --> 0:01:55.520 about the Turing Test, about artificial intelligence, about ways of 0:01:55.560 --> 0:01:59.800 creating piece of software, robots, computer programs, apps, all kinds 0:01:59.800 --> 0:02:03.480 of things that are designed to try to lull you 0:02:03.600 --> 0:02:07.720 into the sense that you're interacting with another conscious agent, 0:02:07.960 --> 0:02:11.959 like a human, even if you're just interacting with the machine. Yeah, 0:02:12.000 --> 0:02:15.400 and that that might have some very sophisticated kind of 0:02:15.680 --> 0:02:18.840 simulation of intelligence, or it might just have a small 0:02:18.880 --> 0:02:21.600 collection of a few clever tricks, right, And it may 0:02:21.639 --> 0:02:25.679 not even be intended to fool you into thinking you're 0:02:25.720 --> 0:02:28.680 talking to a human. It may just be that this 0:02:28.800 --> 0:02:33.680 was an attempt to create a more naturalistic means of 0:02:33.680 --> 0:02:37.720 of addressing an issue, and it wasn't an attempt to 0:02:38.120 --> 0:02:43.000 you know, deceive you. However, there are some UH agents 0:02:43.560 --> 0:02:47.560 that were specifically engineered as a means of deception in 0:02:47.560 --> 0:02:49.640 one way or another. Yea, So today we want to 0:02:49.639 --> 0:02:54.360 talk about bots, specifically bots on the web, what the 0:02:54.400 --> 0:02:58.040 bots are, how to spot them, what some characteristics are, 0:02:58.040 --> 0:03:00.480 and then maybe have a little discussion about what we 0:03:00.520 --> 0:03:04.080 think about the future of bots. Yeah, so we so 0:03:04.160 --> 0:03:06.240 maybe we should start this off by going back to 0:03:06.280 --> 0:03:09.040 an old favorite of ours, which is the Turing test. Right, 0:03:09.080 --> 0:03:11.480 So the Turing test. This was something that was kind 0:03:11.520 --> 0:03:13.520 of proposed by Alan Turing, and it wasn't really a 0:03:13.520 --> 0:03:16.520 test that he was proposing. Yeah, it was more a 0:03:16.880 --> 0:03:19.320 thought experiment. Yeah. He was just kind of saying that 0:03:19.400 --> 0:03:22.480 if you were to design a computer program that could 0:03:22.520 --> 0:03:27.000 fool a human judge a certain percentage of the time. 0:03:27.240 --> 0:03:31.440 Usually it's it's related to being around thirty or third 0:03:31.440 --> 0:03:35.200 of the time, that you could argue that this machine 0:03:35.400 --> 0:03:39.240 is at least simulating the appearance of intelligence. And if 0:03:39.440 --> 0:03:42.400 it can do that reliably, why would you not just 0:03:42.440 --> 0:03:45.320 go ahead and extended the courtesy saying it is intelligent? 0:03:45.400 --> 0:03:48.240 After all, when I have a conversation with another human being, 0:03:48.880 --> 0:03:52.360 I assume that human being is intelligent because of my 0:03:52.440 --> 0:03:55.960 interactions with previous human beings and because of my own experience, 0:03:56.160 --> 0:03:59.240 why would I not extend that same courtesy to a machine. 0:03:59.520 --> 0:04:01.560 Of course, what you might be thinking is the no, 0:04:01.800 --> 0:04:04.520 I don't assume that the humans I interact with our intelligence, 0:04:04.520 --> 0:04:08.000 in which case you're mean, Well, we should observe a 0:04:08.080 --> 0:04:12.040 sort of phenomenal difference between the interactions we typically have 0:04:12.160 --> 0:04:14.960 with humans, or at least the majority of our interactions 0:04:15.000 --> 0:04:18.520 with humans, and the kinds of tests that we now 0:04:18.600 --> 0:04:22.160 would call Turing tests, because typically the Turing test as 0:04:22.160 --> 0:04:26.520 it's imagined today is a text based affair. It's talking 0:04:26.520 --> 0:04:32.960 about communication that's carried out entirely through words written text communication. Right, 0:04:33.120 --> 0:04:35.760 you're not talking about looking at a person and and 0:04:35.760 --> 0:04:38.120 seeing their face. That that's a whole different kind of 0:04:38.160 --> 0:04:41.080 We would probably still call that like life simulation or 0:04:41.160 --> 0:04:45.839 artificial intelligence, like the mimicking of say, human facial movements 0:04:46.040 --> 0:04:48.440 or something like that. And there's been tons of studies 0:04:48.440 --> 0:04:51.520 and robotics to do exactly that, to design robots that 0:04:52.120 --> 0:04:54.680 won't creep people out by having more kind of a 0:04:54.760 --> 0:04:59.279 human interaction and yet having to avoid the uncanny Valley problem, 0:04:59.600 --> 0:05:02.880 where if you get too good actually gets creepy here, right, 0:05:02.880 --> 0:05:06.280 It's it's it's too good and yet not perfect. So 0:05:06.320 --> 0:05:10.080 it's that gap between too good and perfect, are good 0:05:10.120 --> 0:05:13.120 and perfect, where if you make a robot that's good, 0:05:13.279 --> 0:05:15.880 then it doesn't seem creepy. If you make a robot 0:05:15.920 --> 0:05:18.600 that's perfect, it might not seem creepy. No one's done 0:05:18.600 --> 0:05:20.640 that yet, so it's hard to say. But if it's 0:05:20.680 --> 0:05:26.279 really close but not quite perfect, it is unnerving. And 0:05:26.360 --> 0:05:30.280 the Turing test, the various programs that have been said 0:05:30.320 --> 0:05:34.239 to pass the Touring Test in almost every single case, 0:05:34.279 --> 0:05:39.440 we're talking about lowered expectations of what the the bar 0:05:39.600 --> 0:05:43.760 would be, right, right, We're talking about text communication via 0:05:43.880 --> 0:05:48.240 computer with a bot that would purport to be say 0:05:48.320 --> 0:05:52.520 a child whose first language wasn't English, or someone else 0:05:52.520 --> 0:05:58.080 who would be uh, culturally or grammatically hampered in normal conversation, right, 0:05:58.200 --> 0:06:00.440 Or they've just got some clever tricks for changing the 0:06:00.480 --> 0:06:04.200 subject whenever they're not sure that they understand you, right, right, 0:06:04.440 --> 0:06:07.520 for example, purporting to have some kind of psychological condition 0:06:07.560 --> 0:06:11.239 that makes them act really shady. Yeah, that's a yea perfectly. 0:06:11.279 --> 0:06:14.200 All of these are real world examples. Were alluding to right. 0:06:14.240 --> 0:06:17.279 But of course when we talk about these these programs 0:06:17.320 --> 0:06:19.359 that are said to you know, there's there's an article 0:06:19.440 --> 0:06:21.640 you see where they say, oh, we passed the Turing tests. 0:06:21.680 --> 0:06:24.960 These are sort of like staged experiments, right. People set 0:06:25.040 --> 0:06:28.440 up the test on purpose and that they have somebody 0:06:28.480 --> 0:06:31.160 who's trying to determine if they're chatting on some kind 0:06:31.160 --> 0:06:34.760 of instant messenger program with a real person or with 0:06:34.920 --> 0:06:39.760 a computer program. But we actually encounter pretty much this 0:06:39.880 --> 0:06:43.760 scenario in real life in the wild all the time. 0:06:44.200 --> 0:06:47.960 No celebrity judges. Yeah, because bots, let's face it, on 0:06:48.040 --> 0:06:53.279 the web, they are everywhere, and sometimes they are again 0:06:53.320 --> 0:06:56.080 we're aware of it, and it's not a problem. Like 0:06:56.120 --> 0:06:59.880 it might be an automated response to a helpline for 0:07:00.080 --> 0:07:03.080 or some service or product that you use, So you 0:07:03.160 --> 0:07:05.240 might go to a website, and really it's just a 0:07:05.279 --> 0:07:08.479 way of filtering you down so that you either end 0:07:08.560 --> 0:07:11.280 up getting the information you want because it's somewhere buried 0:07:11.280 --> 0:07:14.000 in a help file, but it's way easier if you 0:07:14.120 --> 0:07:17.920 go through a very conversational approach to narrow down what 0:07:18.000 --> 0:07:21.119 the problem is, or it then directs you to the 0:07:21.160 --> 0:07:25.520 person who would presumably have the actual answer you need those, 0:07:25.720 --> 0:07:27.600 that's one type of butt. But we're really kind of 0:07:27.600 --> 0:07:31.720 focusing on the kind that you're not necessarily supposed to 0:07:31.760 --> 0:07:35.160 know was being run by a machine. Right, So I've 0:07:35.160 --> 0:07:39.040 got a question for you, ask away, how much web 0:07:39.080 --> 0:07:43.080 traffic do you think is a real human sitting at 0:07:43.080 --> 0:07:46.440 a browser looking for something because they want or need 0:07:46.480 --> 0:07:49.600 to know. See. If you had asked me this question 0:07:49.800 --> 0:07:53.400 like two weeks ago, I would have said, clearly, all 0:07:53.440 --> 0:07:55.640 of the traffic on the web, aside from maybe like 0:07:55.680 --> 0:07:59.640 a few researchers that like send stuff out to go 0:07:59.760 --> 0:08:02.440 look it stuff just just to see what happens, I 0:08:02.480 --> 0:08:05.600 would have said that like cent of all web traffic 0:08:05.680 --> 0:08:10.160 was actual humans, because why why Joe and Jonathan would 0:08:10.160 --> 0:08:14.760 anyone ever create a robot that looks at websites? Robots 0:08:14.800 --> 0:08:22.120 don't need websites. Pennies, nickels, dollar bills, hundred dollar bills. Yeah, 0:08:22.200 --> 0:08:26.440 we're talking money, money, money. Wow. I didn't expect we 0:08:26.440 --> 0:08:28.400 were going to get to hear that today, Prince Charles, Yeah, 0:08:28.440 --> 0:08:31.160 I busted out when I have to know you got 0:08:31.160 --> 0:08:35.000 to hear these these projected numbers. This is pretty eye 0:08:35.040 --> 0:08:39.480 opening from last year. Yeah, this, uh, there was one 0:08:39.520 --> 0:08:43.120 figure that came from the Interactive and Advertising Bureau that 0:08:43.400 --> 0:08:47.240 suggested that around thirty six of all web traffic is 0:08:47.320 --> 0:08:50.920 quote unquote fake, meaning that it was actual bots, it 0:08:51.000 --> 0:08:56.880 was it was non humans surfing to websites. And uh. 0:08:57.080 --> 0:09:00.360 The reason for this is pretty easy to under stand 0:09:00.360 --> 0:09:04.760 when you realize how a lot of funding for websites happens. 0:09:05.240 --> 0:09:09.480 You have advertising displayed on web pages, and the more 0:09:09.640 --> 0:09:12.280 often people are going to your web page, the more 0:09:12.400 --> 0:09:14.640 valuable that real estate is. It's kind of the same 0:09:14.679 --> 0:09:16.920 as if you want to buy a billboard next to 0:09:17.040 --> 0:09:20.440 the major highway running through your city versus the tiny 0:09:20.480 --> 0:09:24.400 little side street that hardly anyone goes down. The billboard 0:09:24.559 --> 0:09:26.280 next to the highway is probably going to be more 0:09:26.320 --> 0:09:29.000 expensive to use than the one on a little sites, 0:09:29.200 --> 0:09:32.079 and therefore is going to generate more revenue for whoever 0:09:32.120 --> 0:09:35.400 owns the highway or the advertising space right right right, 0:09:35.440 --> 0:09:37.640 So if you own that advertising space, what you should 0:09:37.640 --> 0:09:41.280 actually do is buy up a bunch of autonomous vehicles 0:09:41.320 --> 0:09:43.800 and spend day and night just driving them back and 0:09:43.840 --> 0:09:46.720 forth along that stretch of road. And that sounds crazy, 0:09:46.800 --> 0:09:49.200 but this is what human people are doing right just 0:09:49.320 --> 0:09:51.839 on the Internet, not on actual streets as far as 0:09:51.880 --> 0:09:57.040 you know. So, so this is this is actually a problem. 0:09:57.480 --> 0:10:00.000 This is not This is not just something where you're like, oh, well, 0:10:00.080 --> 0:10:02.959 that's weird that that more than a third of all traffic. 0:10:03.040 --> 0:10:05.560 By the way that the conservative is right, there were 0:10:05.600 --> 0:10:07.640 some experts who are saying that as much as fifty 0:10:07.720 --> 0:10:11.560 percent of traffic on the web was actually generated by 0:10:12.120 --> 0:10:15.520 non human entities. Now, it's worth saying that these probably 0:10:15.559 --> 0:10:18.480 aren't the forward the same forward facing bots you're going 0:10:18.520 --> 0:10:21.959 to interact with the ones that you see in comments 0:10:22.000 --> 0:10:24.240 sections and on social media. We'll get to those in 0:10:24.240 --> 0:10:27.760 a minute, but they're probably not Also on Twitter, bots 0:10:27.760 --> 0:10:31.959 don't usually diversify that. This is essentially algorithms are pretty 0:10:32.000 --> 0:10:35.160 simple exactly. This is. This is an algorithm that that 0:10:35.240 --> 0:10:39.160 navigates a browser to a particular web page, uh and 0:10:39.320 --> 0:10:41.920 stays there for a given length of time. That's another 0:10:41.960 --> 0:10:44.280 thing that you guys may not know is that the 0:10:44.320 --> 0:10:46.920 amount of time spent on a web page can also 0:10:47.720 --> 0:10:51.360 impact how valuable that web pages. So if it's just 0:10:51.400 --> 0:10:53.680 a hit, if that's all they're looking at, well that's 0:10:53.720 --> 0:10:56.000 that's easy. But if it's more than that, if it's 0:10:56.040 --> 0:10:58.199 how much time was spent on that web page, because 0:10:58.520 --> 0:11:00.400 if it was a fraction of a second, then clearly 0:11:00.440 --> 0:11:03.400 they didn't have enough time to view whatever advertisements might 0:11:03.400 --> 0:11:06.640 have been on that page. But as the Bureau has 0:11:06.679 --> 0:11:09.240 pointed out, it is a serious problem because it ends 0:11:09.320 --> 0:11:15.160 up devaluing digital experiences. If advertisers come to a website 0:11:15.160 --> 0:11:18.559 and say, hey, we want to advertise on your page, 0:11:18.559 --> 0:11:20.760 but we happen to know that more than a third 0:11:20.800 --> 0:11:23.959 of all the traffic going to your page is not 0:11:24.080 --> 0:11:26.920 from actual human beings. We're not gonna pay you the 0:11:26.960 --> 0:11:29.160 amount you think this page is worth because it's not 0:11:29.240 --> 0:11:31.199 really worth that. It's not actual human eyes that are 0:11:31.200 --> 0:11:33.520 seeing the stuff we want to put on your page. Right, 0:11:33.600 --> 0:11:36.120 And so I'm not a digital marketing expert, but from 0:11:36.160 --> 0:11:39.679 my understanding, the way this works is it's not typically 0:11:39.720 --> 0:11:42.480 going to be that you know, let's say you operate 0:11:42.520 --> 0:11:45.640 a web page. It's probably not that one third a 0:11:45.640 --> 0:11:49.160 half of your traffic is fraudulent. It's that there are 0:11:49.200 --> 0:11:52.520 a whole lot of sort of spammy, kind of shady 0:11:52.600 --> 0:11:55.560 sites out there that are funneling tons and tons of 0:11:55.600 --> 0:11:59.560 fake traffic to themselves. That's that's sort of changing the average. 0:11:59.600 --> 0:12:01.400 There's either I don't know if you saw this. There's 0:12:01.480 --> 0:12:06.000 even speculation that it can sometimes be a rival advertising 0:12:06.040 --> 0:12:10.880 firm sending huge amounts of traffic to bankrupt their opponent, 0:12:10.920 --> 0:12:14.160 their their competitors by saying, well, our competitor will pay 0:12:14.160 --> 0:12:17.120 out a certain amount of money every time this website 0:12:17.160 --> 0:12:20.800 is viewed. If we flood that website with tons of traffic, 0:12:20.920 --> 0:12:23.319 they're gonna have to pay out tons of money as 0:12:23.360 --> 0:12:25.920 a result because of this agreement they're in, And we'll 0:12:26.040 --> 0:12:30.640 end up hurting our competitor this way, I know. Yeah. 0:12:30.679 --> 0:12:34.600 Another thing that some advertisers do is use bots to 0:12:34.720 --> 0:12:38.920 purchase ads, like software that can identify sites that fit 0:12:38.920 --> 0:12:41.360 within the advertiser's plan and then place ads on those 0:12:41.360 --> 0:12:47.040 sites automatically at high speed. Wow. So sometimes it's bots 0:12:47.160 --> 0:12:52.120 showing ads to bots that were placed by bots. Yeah. Yeah. 0:12:52.240 --> 0:12:56.040 All I can imagine is all the Amazon robots going crazy, 0:12:56.160 --> 0:12:58.360 like we have to deliver no way that the delivery 0:12:58.400 --> 0:13:01.120 was canceled the way we got to deliver, just like 0:13:01.400 --> 0:13:05.760 an insane activity inside the Amazon warehouse. And and according 0:13:05.800 --> 0:13:07.720 to numbers that I read, it's only about four per 0:13:07.800 --> 0:13:11.199 cent of bots that I purchased that of ads that 0:13:11.240 --> 0:13:14.840 are purchased by bots as of but media marketing folks 0:13:14.840 --> 0:13:17.760 think that that could boom to in the next ten years, 0:13:17.840 --> 0:13:21.280 that it is an enormous growth. I think that might 0:13:21.320 --> 0:13:24.280 be helpful. I'm not sure. So if you are like 0:13:24.320 --> 0:13:27.920 an advertiser trying to advertise on a website and you're 0:13:27.920 --> 0:13:32.360 trying to figure out if it's traffic is real or fake. Again, 0:13:32.400 --> 0:13:34.480 from what I understand, in a lot of cases, this 0:13:34.520 --> 0:13:36.760 is pretty easy to do. Like you can look at 0:13:36.800 --> 0:13:39.959 some certain metrics and kind of say, okay, well, if 0:13:39.960 --> 0:13:42.840 you have the access to the metrics, right, yeah. See, 0:13:42.840 --> 0:13:45.800 That's the thing is that actually it behooves the the 0:13:45.840 --> 0:13:48.520 people who are running the website to keep an eye 0:13:48.559 --> 0:13:51.320 on those metrics to make sure that they are in 0:13:51.440 --> 0:13:56.360 fact legitimate, because if it were ever discovered that they aren't, 0:13:56.440 --> 0:14:00.320 whether whether ultimately the owner of that website was at fault, 0:14:00.559 --> 0:14:03.560 it's going to negatively impact that owner. Right, So, if 0:14:03.600 --> 0:14:07.920 the owner is employing bots essentially to visit the website 0:14:07.960 --> 0:14:11.480 to drive up numbers, if that ever gets public, that's 0:14:11.480 --> 0:14:14.960 a huge black mark against that that website and that owner. 0:14:15.440 --> 0:14:17.360 If in fact someone else is doing it and the 0:14:17.400 --> 0:14:20.280 web owner is completely innocent of it, it can still 0:14:20.360 --> 0:14:22.680 be a black mark. So it's one of those things 0:14:22.720 --> 0:14:26.720 that now, honestly, if you really look at the long term, 0:14:26.840 --> 0:14:30.000 you don't want this happening because it ends up hurting 0:14:30.080 --> 0:14:32.680 everybody in the long run. It's only the short term 0:14:32.680 --> 0:14:35.240 where it seems to have some gains. Right. But let's 0:14:35.280 --> 0:14:37.200 if you are one of these people who's trying to 0:14:37.320 --> 0:14:40.160 use one of these sneaky tricks to like generate a 0:14:40.160 --> 0:14:45.520 bunch of fake traffic. Obviously, the artificial intelligence capabilities are 0:14:45.560 --> 0:14:48.720 getting better, and this is an interesting variant on the 0:14:48.760 --> 0:14:52.280 type of artificial intelligence we usually talk about. Usually we 0:14:52.360 --> 0:14:57.720 talk about a kind of um single agent oriented artificial 0:14:57.760 --> 0:15:01.000 intelligence simulation, Like when you're doing our official intelligence, you're 0:15:01.000 --> 0:15:03.520 trying to create a program that can talk like a 0:15:03.520 --> 0:15:08.040 single person or behave like a single person. This would 0:15:08.080 --> 0:15:13.760 be artificial swarm intelligence, like it's trying to behave the 0:15:13.800 --> 0:15:18.880 way massive groups of people actually look in statistics. Yeah, 0:15:18.880 --> 0:15:20.960 and you can imagine, like, if you want to get 0:15:21.000 --> 0:15:23.800 super clever with this one, you could try and design 0:15:23.840 --> 0:15:27.480 an algorithm that would surf to a page as if 0:15:27.560 --> 0:15:30.240 it had been human, in other words, to follow a 0:15:30.320 --> 0:15:34.240 trail as opposed to just directly navigate to a site 0:15:34.680 --> 0:15:36.960 so that if someone were looking into it, it would 0:15:36.960 --> 0:15:39.560 look as if it was an actual person. And if 0:15:39.600 --> 0:15:42.840 you really wanted to be tricky, you could look for 0:15:43.120 --> 0:15:49.520 times when it's the most advantageous to initiate a swarm 0:15:49.560 --> 0:15:52.640 activity at a website. For example, let's say that there 0:15:52.680 --> 0:15:56.240 was a big press release. Then that ends up naturally 0:15:56.240 --> 0:15:59.840 saying well, curiosity would be raised because we've they've got 0:15:59.840 --> 0:16:02.920 some thing new to announce. Therefore you would expect an 0:16:03.000 --> 0:16:07.920 increase in traffic. We're just artificially boosting that increase. So 0:16:08.200 --> 0:16:12.360 it gets super like super secret spy type stuff going on. 0:16:12.440 --> 0:16:14.680 And this is all on the corporate end, right. But 0:16:14.800 --> 0:16:18.200 let's say you're not involved in any kind of web 0:16:18.280 --> 0:16:21.960 advertising or anything like that. You're just a normal Internet user. 0:16:22.040 --> 0:16:25.160 You use some social media occasionally, you look at the comments. 0:16:25.200 --> 0:16:27.280 You know you shouldn't, but you do. Yeah. I'm pretty 0:16:27.360 --> 0:16:29.280 much with you, except for the normal parts. So go ahead. 0:16:31.880 --> 0:16:34.600 Yeah I haven't figured it out. Well, it turns out 0:16:34.720 --> 0:16:39.320 normal is encountering tons of bots all the time. Yeah, 0:16:39.480 --> 0:16:42.280 we are all going to be encountering lots of bots 0:16:42.800 --> 0:16:45.360 all the time. So you're gonna find bots on Twitter 0:16:45.800 --> 0:16:48.640 if you use online dating, you're going to find bots 0:16:48.680 --> 0:16:51.880 on their trying to scam you body accounts. Uh. There 0:16:52.000 --> 0:16:55.640 is their bots on Facebook. Their bots in the comments 0:16:55.640 --> 0:17:00.640 section of almost any article or anything you look at. Yeah, 0:17:00.680 --> 0:17:04.080 they're everywhere. Whenever you see a like you go to 0:17:04.240 --> 0:17:07.360 an article or a video, anything really that allow has 0:17:07.359 --> 0:17:09.919 a comment section. And whenever you see someone who is 0:17:09.960 --> 0:17:13.439 posting a link to something that is truly unrelated to 0:17:13.720 --> 0:17:16.520 whatever the content is, that's more likely than not a 0:17:16.520 --> 0:17:19.520 bot account. Sure, And even if it's posting a link 0:17:19.560 --> 0:17:22.480 to something that is related, that can still be about account. 0:17:22.600 --> 0:17:25.679 And even if it's just seeming to participate in the 0:17:25.720 --> 0:17:31.000 conversation like a human would, it can be about account. Yeah, yeah, 0:17:31.040 --> 0:17:34.280 it can be. Yeah, we're going from the the easy 0:17:34.400 --> 0:17:38.520 to detect to the more difficult to detect. Sometimes, depending 0:17:38.560 --> 0:17:42.679 upon the interaction, it may be really hard to detect. 0:17:42.720 --> 0:17:45.879 And we'll get into that. Okay, So just for the record, 0:17:46.119 --> 0:17:48.640 what do we mean when we say butt, Like, what 0:17:48.640 --> 0:17:51.920 what distinguishes a bot from you know, a full fledged 0:17:52.000 --> 0:17:56.720 artificial intelligence program or something. Complexity is a big part 0:17:56.720 --> 0:17:59.720 of it, right for for a truly like depending upon 0:17:59.800 --> 0:18:02.640 why you're talking about, when you say AI, because, as 0:18:02.680 --> 0:18:07.080 we've discussed before, AI covers a wide variety of of 0:18:07.560 --> 0:18:10.600 disciplined behaviors, that sort of thing like you could argue 0:18:10.680 --> 0:18:15.200 that having a sensor that sends a message to a 0:18:15.240 --> 0:18:19.640 program that then has a reaction based upon the information 0:18:19.680 --> 0:18:22.680 the sensor is gathered, that's the type of AI. But 0:18:22.680 --> 0:18:24.440 it probably it might be a reach. You could even 0:18:24.520 --> 0:18:27.879 argue that, like a robot that successfully squirts ketch up 0:18:27.960 --> 0:18:29.720 on your plate or on your hot dog has some 0:18:29.800 --> 0:18:31.760 kind of well yeah, because it has to detect where 0:18:31.800 --> 0:18:33.639 the hot dog is and then detect you know, how 0:18:33.680 --> 0:18:36.400 much pressure it needs to use. There is some there's 0:18:36.440 --> 0:18:39.080 intelligence involved in catch up. Yeah. No, if you ever 0:18:39.119 --> 0:18:40.840 see the way that I end up trying to put 0:18:40.880 --> 0:18:43.399 ketch up on something, you realize I lack that intelligence. 0:18:47.119 --> 0:18:51.159 It does we I thought we agreed we weren't going 0:18:51.200 --> 0:18:55.360 to talk about the lunch meeting at the podcast, but yeah, so. 0:18:55.600 --> 0:18:57.360 But at any rate, if we're talking about the type 0:18:57.400 --> 0:19:01.840 of artificial intelligence that is meant to maintain a prolonged 0:19:01.880 --> 0:19:05.200 conversation or the type that is supposed to at least 0:19:05.400 --> 0:19:09.600 simulate UH intelligence on a more general level, what we 0:19:09.640 --> 0:19:13.280 would start to look at as potentially being strong AI, 0:19:13.400 --> 0:19:16.480 or at least simulating the appearance of what we would 0:19:16.480 --> 0:19:19.360 consider strong AI. Strong AI being a machine that can 0:19:19.400 --> 0:19:23.440 truly think, right, a machine that is capable of thinking 0:19:23.480 --> 0:19:28.840 and making decisions, perhaps even having consciousness, Although that there's 0:19:28.840 --> 0:19:31.000 an argument over whether or not that would be necessary 0:19:31.040 --> 0:19:34.200 to truly be strong AI. We're still talking about weak AI. 0:19:34.240 --> 0:19:36.439 In that sense. Bots can be even simpler than that. 0:19:37.000 --> 0:19:42.119 Bots can be so simple as to really be uh, 0:19:42.200 --> 0:19:45.400 you know, closer to that that model of a sensor 0:19:45.480 --> 0:19:48.720 that sends information and then the computer reacts in a 0:19:48.760 --> 0:19:51.800 specific way. It may be as simple as that. Yeah. 0:19:51.800 --> 0:19:54.879 Typically it's thought of as something that's very simple. It 0:19:54.960 --> 0:19:58.919 automates a task on the internet. So you can have 0:19:59.000 --> 0:20:01.480 like hacker bots a that's another kind of thing that 0:20:01.520 --> 0:20:03.440 we're probably not really going to talk about today. If 0:20:03.480 --> 0:20:06.320 somebody might create a bot army of a bunch of 0:20:06.359 --> 0:20:10.919 computers with with programs secretly installed the hijack those computers 0:20:10.960 --> 0:20:13.639 to do what the hacker wants, and some of the 0:20:13.680 --> 0:20:15.879 butts we're talking about here may in fact be the 0:20:15.960 --> 0:20:20.120 result of that, because it means that the person who 0:20:20.200 --> 0:20:24.160 is ultimately responsible for creating the butts has a level 0:20:24.200 --> 0:20:27.399 of protection because if it gets traced back, it's traced 0:20:27.440 --> 0:20:30.760 back to the victims computer, not to the hackers computer. Right. 0:20:30.840 --> 0:20:33.680 But then also a lot of bots on the web 0:20:33.720 --> 0:20:37.960 that we interact with our they're so simple that they're 0:20:38.000 --> 0:20:42.399 not even really like artificial intelligence. I mean, they're they're 0:20:42.520 --> 0:20:45.680 a collection of tricks. They might have like three tricks, 0:20:46.400 --> 0:20:48.920 and they just do them over and over and most 0:20:48.960 --> 0:20:50.919 of the time they're not going to fool anybody, but 0:20:51.000 --> 0:20:53.639 every now and then they do, and they're so cheap 0:20:54.400 --> 0:20:56.960 you might as well use them. It's a shotgun approach 0:20:57.320 --> 0:20:59.680 of just like, well, these aren't going to fool everyone 0:20:59.720 --> 0:21:02.480 all time, obviously, but like if that one or two 0:21:02.520 --> 0:21:07.280 get through, then same same same philosophy strategy as as 0:21:07.320 --> 0:21:10.200 people who send out or create networks to send out 0:21:10.200 --> 0:21:14.879 spam emails, right, the idea being that, sure, the people 0:21:15.200 --> 0:21:18.200 who get this are never going to respond to it. Yeah, yeah, 0:21:18.200 --> 0:21:20.639 they're not going to say, I believe you Nigerian Prince, 0:21:20.720 --> 0:21:23.800 thank you, thank you for this information exactly. That one 0:21:23.840 --> 0:21:26.560 percent that does respond, however, if you're sending it to 0:21:26.640 --> 0:21:29.520 a large enough group of people is still a significant number. 0:21:30.080 --> 0:21:32.480 So it's the same kind of strategy there. In fact, 0:21:32.560 --> 0:21:35.080 if we want to give an example to explain this 0:21:35.200 --> 0:21:38.679 simple behavior, you could use Twitter. It's a very easy 0:21:38.760 --> 0:21:41.679 platform to describe. Let's say that you set up an 0:21:41.680 --> 0:21:45.320 algorithm that does a search for certain keywords. Whenever those 0:21:45.400 --> 0:21:49.000 keywords pop up, it identifies the person that sent the 0:21:49.040 --> 0:21:52.800 tweet that contained those keywords and sends an apt reply 0:21:52.960 --> 0:21:56.840 to that person with a link attached. Uh, the goal 0:21:56.920 --> 0:22:00.800 being for that person to end up clicking through that link. Hey, 0:22:00.840 --> 0:22:03.199 you mentioned lunch, are you interested in weight loss? Right 0:22:03.280 --> 0:22:05.399 through to the spam page? Right? And it may it 0:22:05.440 --> 0:22:09.840 may even be more more um uh, you know, subtle 0:22:09.920 --> 0:22:12.800 than that. It maybe like I think you meant and 0:22:12.840 --> 0:22:16.680 then the link or that's cool, have you seen this 0:22:16.960 --> 0:22:19.120 and then the link? You know, something so that it's 0:22:19.160 --> 0:22:23.359 harder for you to initially identify that as a sales pitch. Yeah, 0:22:23.520 --> 0:22:26.360 a lot of them work by i would say, limiting 0:22:26.440 --> 0:22:30.600 interaction or information. Yeah, it's just it's a pretty simple trick, 0:22:30.680 --> 0:22:33.199 but sometimes it works. Another thing is I think a 0:22:33.200 --> 0:22:37.000 lot of times Twitter bots overcome their simplicity by hiding 0:22:37.000 --> 0:22:40.200 in plain sight, like they don't seem to care all 0:22:40.240 --> 0:22:42.920 that much if they look like a real human they're 0:22:42.960 --> 0:22:45.680 just enough of them working in tandem that they can 0:22:45.760 --> 0:22:48.480 blend in with a mixed crowd of real humans and 0:22:48.520 --> 0:22:52.440 other bots, especially for certain kinds of uses. And one 0:22:52.440 --> 0:22:56.960 of those that I would identify is swaying public opinion. Sure, 0:22:58.119 --> 0:23:01.000 so have you all heard about propaga into bots? Yeah, 0:23:01.080 --> 0:23:07.000 I've heard about these, and uh it's incredibly upsetting. So 0:23:07.280 --> 0:23:11.199 there are political bots out there, It's a phenomenon on 0:23:11.600 --> 0:23:15.600