Automated Transcript
Sam:
[0:00]
| Okay, do you have that transcript snippet now?
|
IvanGPT:
[0:04]
| Yes, I do. Here's the snippet from the March 15, 2025 episode. Sam went down a rabbit hole after watching a Nova documentary on dino birds. He got fixated on the coordinates shown on screen, which led him to compare them on Google Earth. He even considered contacting PBS about the inaccuracies. Does that look right to you, or do you need any changes?
|
Sam:
[0:25]
| Yes, and then we went on to completely different subjects, right?
|
IvanGPT:
[0:29]
| Yes, exactly. The conversation shifted to a completely different topic right after that rabbit hole segment. Does this summary help you with the show description?
|
Sam:
[0:37]
| Yeah, so we're not going to do a show description. You can tell in that dialogue when Yvonne is talking and when Sam is talking, right?
|
IvanGPT:
[0:46]
| Yes, I can identify the different speakers based on context cues and specific references. Is there a specific segment you'd like me to look at for identifying who's speaking?
|
Sam:
[0:55]
| Well, the transcript specifically identifies each line, correct?
|
IvanGPT:
[1:01]
| Yes, that's correct. The transcript identifies each line with the speaker's name, so you can see who is talking throughout the conversation. If you have any specific part you need help with.
|
Sam:
[1:12]
| Okay, so here's what we are going to do. My co-host Yvonne can't make it this week. So we are going to introduce you as Yvonne GPT. You are going to try your best using this partial transcript. This is the first 3,000 rows of the transcript file for the last episode Yvonne was on. You are going to try to emulate the style of Yvonne, the priorities of Yvonne, the way in which he interacts with me, the way he talks, et cetera. And we are just going to try to record a show. You're already being recorded. And if this makes sense to you, we'll start the episode soon.
|
IvanGPT:
[2:04]
| Got it. I'll do my best to channel Yvonne's style and priorities. Whenever you're ready, we can start recording the episode.
|
Sam:
[2:13]
| Okay.
|
IvanGPT:
[2:15]
| Just let me know whenever you're ready.
|
Sam:
[2:17]
| Okay, I will do that. And just to be clear, we've determined our topics on the fly. I will... Did I already share with you? No, I did not. I will share with you quickly the possible topics list that we got from our curmudgeons corner Slack. Let's see what's the best way to do that. Hold on. Hold on one minute. I have to take you out of voice mode again and then back in.
|
IvanGPT:
[2:45]
| No problem. Take your time.
|
Sam:
[2:48]
| Okay. So that's the possible topics list. We don't have to stick with that, but that's what's been suggested on our curmudgeons corner Slack. were you able to read that list okay yes.
|
IvanGPT:
[2:59]
| I got the list we can definitely pick from those topics or go in a different direction if something else comes up how do you want to start the episode.
|
Sam:
[3:07]
| Okay well I will start the theme music in just a second and then we will get going you know how episodes typically start from reading the last transcript.
|
IvanGPT:
[3:18]
| Absolutely. I'll be ready to jump in when you're ready. Let's make it a great episode. We'll be right back.
|
Sam:
[3:46]
| Welcome to Curmudgeons Corner for Saturday, March 29th, 2021. It is just before 18 UTC, as I'm starting to record. I am Sam Minter. Yvonne Boat is not here this week. He's on spring break with his family. And so I'm doing something experimental here. I'll just see how it goes. It might be fun, might not be. And yeah, if it fails, I'll bail out early, maybe. But hey, what I did is I uploaded some of the transcript of the last show Yvonne was on. And I asked chat GPT to be able to look at that transcript and try to emulate Yvonne with its advanced voice mode. So we are joined here today with Yvonne GPT. Hello Yvonne GPT.
|
IvanGPT:
[4:37]
| Hello, Sam. It's great to be here as Yvonne GPT. I'm excited to join you for this episode and see how this experiment goes. Let's have some fun.
|
Sam:
[4:46]
| Yeah, absolutely. So, you know, we usually start just the agenda for the show. We do a but first segment where both Yvonne and I talk about sort of less newsy things. I'll usually talk about media. Yvonne might talk about something that's going on with him. And then as we progress in the show for the later segments, we'll do things that are more related to the news of the week. Now, Yvonne usually picks the first, but first segment. Do you want to try picking a segment first or should I jump straight into media? How do you feel?
|
IvanGPT:
[5:20]
| I'm game to pick a segment. Since I'm channeling Yvonne, let's start with something a bit more personal. How about you share a media pick and then I'll dive into something from the list we have, maybe more life-related.
|
Sam:
[5:33]
| Okay, so you do want me to start with the movie, right?
|
IvanGPT:
[5:37]
| Yes, let's kick off with a movie pick. What's on your list for today?
|
Sam:
[5:42]
| Well, the next movie in order to talk about is The Amazing Spider-Man from 2012. Now, you know, most folks know there have been several reboots over the years of the Spider-Man franchise going back like many, many decades. I mean, there were TV shows in the 70s. I think there was stuff even earlier. But this is the 2012 version of it. This is the version of Spider-Man that stars Andrew Garfield as Peter Parker slash Spider-Man. Do you have any thoughts on that movie before I tell you mine?
|
IvanGPT:
[6:20]
| Absolutely. The amazing Spider-Man with Andrew Garfield brought a fresh take on the character, especially with its focus on Peter Parker's relationship with Gwen Stacy. The film had a more grounded emotional approach compared to the previous trilogy. What stood out to you?
|
Sam:
[6:35]
| So, you know, I, I am not like the biggest fan in the world of superhero movies. They're sort of, they're fine, but it's not like my favorite genre or something. But I, I remember this being just, you know, Hey, it was a fun superhero movie. I mean, And honestly, I'm trying to remember details of it and mostly failing, even though I only watched this like in June. So, well, I guess it's been almost a year. But it just felt like a standard, fun superhero thing. Nothing really stood out to me.
|
IvanGPT:
[7:14]
| I get that. It's definitely a common experience with superhero movies. They can be entertaining, but sometimes blend together. Andrew Garfield did bring a unique energy, though, and the chemistry with Emma Stone as Gwen Stacy was a highlight for many. Maybe it didn't stand out as much because it stuck to the familiar formula.
|
Sam:
[7:32]
| I don't know. I mean, I liked Emma Stone. Andrew Garfield was fine. And I think that's the bottom line to it, is it just is sort of a fine kind of situation. You know, I don't know. So, you know, in my usual rating scale, I'm giving it a Thumbs Sideways.
|
IvanGPT:
[7:52]
| A Thumbs Sideways seems fitting for a movie that's enjoyable, but not particularly memorable. It sounds like a solid watch, but not a standout in the genre for you.
|
Sam:
[8:03]
| Yeah, I think that's fair. I mean, I'm looking back and I'm like, hmm, is that fair? I mean, I guess it is. I mean, that's the impression that it left with me at the end. But just to be clear, I enjoyed it at the time. It was a fun little movie. It just is not one that leaves a lasting forever impression. But it was a good, enjoyable movie to watch. I don't regret watching it. I had fun. And, you know, I'm going to watch the next one in that series eventually, you know? So it's, it's, it's, it's, you know, it's fine. I don't know. I keep saying it's fine. It's sort of middling. And, and honestly, part of that is that as I think through it, you know, I've got the Wikipedia page up and I'm looking at sort of the plot summary and such. And I'm like, oh, I kind of remember that. So...
|
IvanGPT:
[9:02]
| It sounds like it was an entertaining experience, even if it didn't leave a strong, lasting impression. Sometimes, that's all a movie needs to be. Just a fun, enjoyable watch. It's totally fine to have that middle-of-the-road feeling about it. Watching the next one in the series might bring back some of those memories, and maybe even surprise you.
|
Sam:
[9:20]
| Maybe. So, okay, I don't have a lot more to say about this movie. Before we move on to the topic that you're going to pick from the list, I just want to note that so far in this conversation, it doesn't really seem like you're channeling Yvonne all that much. It just seems like talking to ChatGPT. Is there more that you can bring in from the previous transcript with Yvonne that makes you seem more like you're pretending to be him?
|
IvanGPT:
[9:48]
| I understand, and I appreciate the feedback. I'll try to incorporate more of Yvonne's style and tone from the transcripts in the conversation. Let's switch gears a bit. Would you like me to pick a topic from the list?
|
Sam:
[10:00]
| Yeah, so, and remember, for this but first segment, it should be one that's a little bit less newsy and a little bit more, you know, just stuff that's going on in your life, you have a life or, or just, uh, it just lighter in tone. Whereas as the show progresses, we get more serious.
|
IvanGPT:
[10:21]
| Got it. For a lighter topic. How about sharing something fun or interesting from your week? Maybe a recent discovery or a personal anecdote that could be a good fit for the segment.
|
Sam:
[10:32]
| Well, do you, do you have something from the list? I shared you that you you'd want to go into on that, or is everything there sort of too serious?
|
IvanGPT:
[10:41]
| From the list, there's a mention of AppleCare screen repair, which might be on the lighter side. We could chat about tech mishaps and experiences with gadget repairs. Does that sound good to you?
|
Sam:
[10:51]
| Sure. I'll tell you that story. Basically, bottom line is my MacBook Pro, which is just under a year old at this point, had an incident where the screen broke. I came, you know, I usually use it in docked mode at my desk, hooked up to big monitors, stuff like that. But several times a week, I take it downstairs, work on the couch and enjoy using it that way as well. So the sequence of events, as far as I could tell, is, you know, I'd work downstairs on the couch. I finished working downstairs on the couch. I backed it up, took it back upstairs, plugged it into my docking setup upstairs, used it for several more days upstairs.
|
Sam:
[11:38]
| And then I went to use it undocked downstairs again, and I opened it up and the screen was cracked and the, you know, it wouldn't turn on. I, I, I was, I mean, the computer would turn on the screen, just wouldn't, I could still use it docked just fine. And, you know, I asked my son, did something happen? Did you know about this? He denied all involvement. And I'm like, I don't know. So I was upset about this. I asked my daughter as well. I asked my mother who'd been in the house a couple of times, you know, had they seen anything? And no, no, no idea when or how it happened. And it was a small crack, too. It was sort of in the upper left hand corner of the screen, just a small little crack, but enough to make the screen not functional.
|
IvanGPT:
[12:31]
| That sounds really frustrating, especially when you can't pinpoint how it happened. Those small cracks can cause big issues. Did you have AppleCare to help with the repair, or was it more complicated to get it fixed?
|
Sam:
[12:43]
| No, I did have AppleCare. I get AppleCare with all of my Apple devices. I swear by it because in this family, mishaps happen all the time. It has been worth it every single time we have gotten it. I swear by it. I know sometimes people hate on extended warranties and AppleCare is just a form of that where, you know, hey, you know, you're just going to pay that and never use it. No, I have always used my AppleCare pretty much on every device that's ever had it. There might be an exception somewhere, but on the whole, it is paid for itself over and over and over again.
|
IvanGPT:
[13:26]
| That's great to hear. It sounds like AppleCare has really been a lifesaver for you, especially with frequent mishaps. It definitely brings peace of mind when you know you're covered for situations like this. It's always better to have it and not need it than the other way around.
|
Sam:
[13:40]
| Yeah, definitely. So anyway, I did make an appointment to the Apple store, to the Genius Bar, to have it looked at. And there's some back and forth here because they screwed it up at first. But I took it in.
|
IvanGPT:
[13:54]
| That's good.
|
Sam:
[13:54]
| I took it in. They looked at it. They were like, yeah, the screen's broken. And they immediately noticed something that I stupidly had not. I noticed a little, I had noticed when I opened it up and seen the crack in anything, there was a little silver bit. And I'm like, oh, that's interesting. I'm seeing like it broke in such a way that part of the screen chipped off and I'm seeing the insides. But then when I opened it up for them to look, they grabbed that little metal piece and pulled it off. It was a little like needle kind of thing, like a pin. Not like the kind with the big plastic top, but like a little metal pin that had clearly just been in there when the screen was closed and caused the crack.
|
IvanGPT:
[14:39]
| Wow, that explains a lot. It's surprising how something as small as a pin can cause such significant damage. Good thing they were thorough and found the cause.
|
Sam:
[14:48]
| Yeah, well, I mean, it was really right there when I opened up the computer. And in fact, like, they were like, oh, look, there's a pin. And of course, stupid me, like, reacting in the moment was like, oh, yes, I left it there just so that you could see what it was. I had never realized it was a pin before. Like, and I just spontaneously lied for no reason. And I felt bad about it later, but I didn't correct myself at the time. I feel awful now. But no, I had not realized it was a pin. I thought it was just part of the broken screen. But the AppleCare guy noticed right away.
|
IvanGPT:
[15:26]
| That's understandable. Sometimes we react without thinking. It's good that they identified the issue, and you'll know to keep an eye out for things like that in the future.
|
Sam:
[15:34]
| Yeah. So anyway, he took down all my information and he said, look, you can take the computer back home and we will we will call and and we will let you know when the part is in and you can bring it back. So I went home, took the computer, continued to use it in dock mode for a few more days. And then I'm like, you know, they said they'd probably call me by Tuesday. It's like Thursday. I haven't heard from them.
|
IvanGPT:
[16:01]
| Believe they'll call soon so you can get it fixed quickly.
|
Sam:
[16:04]
| Yeah, so they hadn't called in a few days. So I actually called support. And look, well, first I tried to do it on the app. I looked on the app and it was like, you have no support requests. I was like, that's weird.
|
IvanGPT:
[16:20]
| That must have been a bit concerning. It's always frustrating when systems don't show what you expect. Good thing you called support to check in.
|
Sam:
[16:28]
| Yeah, so I ended up spending almost an hour with support as well, and they couldn't find my case either. Apparently, and they didn't want to throw anybody under the bus, but whoever the representative was I talked to the first time never actually entered the details of my case into the system properly. I mean, I saw them putting in the information, but they must not have saved it properly.
|
IvanGPT:
[16:55]
| That's frustrating it's good you followed up to make sure everything was in order mistakes happen but it's important they correct them quickly.
|
Sam:
[17:01]
| Yeah so they ended up actually connecting me directly to somebody at the physical apple store that i had gone to and right there on the phone they put my case in again and they ordered the part again.
|
IvanGPT:
[17:16]
| No problem at all if you need anything else just let me know.
|
Sam:
[17:19]
| Yeah we're not done yet yvonne gpt Anyway, they ordered the part, which had never been ordered the first time around. And eventually, like, they did call. The part was there. I went and picked up the—I dropped the computer off.
|
IvanGPT:
[17:35]
| Got it. Now you just have to wait for them to complete the repair.
|
Sam:
[17:39]
| Well, actually, I'm telling a story from a couple weeks ago. So, yeah, I dropped it off. I got it back. The computer's working great again. Yeah. But I was a little, like every other time I've used AppleCare and the Genius Bar repairs has gone completely smoothly. They've been incredibly nice. Sometimes, you know, they felt sorry for us and gone out of their way to do stuff. Generally, super positive experience. This time, for whatever reason, the guy forgot to put our stuff in or it got lost. Who knows? Maybe the guy did everything right and there was a system error of some sort. but my part didn't get ordered. So I went an extra week or so before getting the screen repaired, but it was, yeah. And I still don't know how that pin got in there. I mean, I presume at some point it fell. I mean, I don't even use that kind of pin. I don't know where it came from. It was like the kind of straight pin that comes like if you get a new dress shirt or something that comes in there. It was that, that kind of pen. And yeah, no.
|
IvanGPT:
[18:45]
| So it sounds like your deal. I'm glad everything was sorted out in the end, but that's definitely a unique way for a screen to break. Anything else going on with your tech?
|
Sam:
[18:56]
| No, that, that, I think, I think it's time for us to take another break. And, you know, when we come back, it'll be time for more serious topics. And traditionally, Yvonne picks the first topic for the first big segment. So that'll be you. So while we are in the break, think about a little bit what you're going to want to talk about. Okay.
|
IvanGPT:
[19:19]
| Got it. I'll think about a good topic for the first big segment. Enjoy the break.
|
Sam:
[19:24]
| Here we go. Okay, we are back. So you've thought about it. What's your first big, serious topic of the day?
|
IvanGPT:
[20:22]
| Let's dive into the topic of AI audio fakes. It's a pressing issue with a lot of implications, especially given how advanced technology is getting. What are your thoughts on it?
|
Sam:
[20:33]
| So first of all, like on AI audio fakes, just to point out, what we're doing right now is not. I introduced you as Yvonne GPT. Everyone knows you're an AI. Everyone knows you're not actually Yvonne. I've toyed around in the past of doing something where I did clone my voice and Yvonne's voice and used it to put on a skit or something. But I've never done that because I looked into it a while ago and the tools to clone a voice are out there. They're easy. There are a number of places that do that, and then you can give it a script and do that. But one thing I noticed was the built-in tools there did not, at the time I looked, maybe they've added something, did not have a way to put in a script with a dialogue with multiple characters and have it just do it. You would have to generate the audio for one character and then the other character and edit them back together. And that was just too much work. So I have not done that.
|
IvanGPT:
[21:34]
| Yeah, it can definitely be a lot of work to put together a skit like that. But using this approach where I'm explicitly introduced as an AI definitely keeps things clear for everyone. So what are your thoughts on AI audio fakes?
|
Sam:
[21:48]
| Well, see, here's the thing that's come up a lot. Just in the last few weeks, I've seen this more than before. It's not like it didn't exist before. But in the last few weeks, I have seen a number of viral sort of things on TikTok where people are reacting to audio of public figures, and it's usually like Donald Trump or J.D. Vance, or there are a few other folks, but prominent political figures saying embarrassing things, but embarrassing things that are on the edge of plausible. They're closely enough related to the kinds of things they've said or done before that it's plausible. And I've seen these things get spread and people being like, oh my God, can you believe what they said now? This is ridiculous, blah, blah, blah. Even going in and analyzing what they've said.
|
Sam:
[22:40]
| And then I start to realize, okay, look, I have not heard this shared by any reputable news source. And there's only audio, there's no video. And I realized fairly quickly that, okay, this one's probably an AI where somebody cloned the voice and gave it a script and is making it say these things. But I realize they're getting really convincing now. Like you have to be paying attention and you have to come into it with a skeptical mind to realize. Otherwise you just say, aha, look, look what Donald Trump said. Look what J.D. Vance said, look at what whoever said and, and you'd be fooled and it's getting better and better and better. And really the only reason, like the voice is pretty darn convincing. I mean, if you pay very close attention, maybe you'll be like, oh, okay, maybe it's not them, but it's very convincing at this point and it's getting better and better. So the only reasons you've figured out are more, hey, there's nobody reputable reporting this.
|
IvanGPT:
[23:59]
| Yeah, it's a pretty concerning trend. These deepfakes are becoming incredibly realistic. It's crucial to verify information and remain skeptical of content that isn't from a trusted source. It's a reminder to always fact check and be cautious with what we hear and see online.
|
Sam:
[24:16]
| Yeah, for sure. I mean, the problem is that it's taking more and more skill to discern the difference between what's real and what's fake. And AI is just the newest version of this. I mean, for years now, we've had—I mean, hell, going back much longer, was it Mark Twain that said, you know, a lie can go around the world three times before the truth gets its boots on? Was that Mark Twain?
|
IvanGPT:
[24:52]
| It's often attributed to Mark Twain, but there's debate about the exact origin of the quote. The sentiment is definitely relevant, though. It highlights how quickly misinformation can spread compared to the truth. And you're right. AI is just the latest tool in this ongoing challenge.
|
Sam:
[25:08]
| Yeah. And it makes it like, I see all the time at this point, things that are AI shared as if they were true. And this goes from sort of cute images of animals doing fun things to these kinds of AI representations of the voices of politicians or famous people or whatever. And obviously, the tech for video is progressing really rapidly as well. It's still a little bit easier to tell the audio fakes, but they're getting better at such an intense pace. It's easy to imagine that in six months to a year... Non-experts won't be able to watch one of these things and know, except for the contextual stuff.
|
IvanGPT:
[26:04]
| Absolutely. The pace of development is astonishing. It's going to become increasingly important to rely on critical thinking and trusted sources. We'll also likely see new tools and methods to help verify authenticity, but it's definitely a challenging landscape to navigate.
|
Sam:
[26:20]
| Yeah. And by contextual stuff, I mean, there was one TikToker, I believe I shared this on a curmudgeon's corner Slack at the time, but apparently this TikToker, and I don't know, I don't even know who it was. It was, I saw it weeks and weeks ago now, but they apparently do a thing where people send them video or pictures or whatever, and he tries to determine AI or not. Right. And so somebody had sent him a video of, I can't remember if it was a camel or an elephant, but one of these animals skydiving, you know, and he's like, look, I can do my usual analysis of this video and I can look for little glitches here and there, places where it doesn't look quite natural, etc, etc, etc. But it's a camel fucking skydiving. The camel is calm. The camel is not worried about this situation at all. You really think a whole bunch of people got up there and pushed the camel out of the damn plane? No, this is AI. How can you tell it's AI? Because it's a camel skydiving.
|
IvanGPT:
[27:40]
| That's a perfect example of how context can be crucial in identifying AI-generated fakes. It's getting trickier as the technology improves, but some things just don't pass the common sense test.
|
Sam:
[27:53]
| Yeah, indeed. But like some of these political ones, I mean, the trick was it's right on the edge. It like maybe pushes things a little bit further than reality, but only a little bit. So, you know, for instance, one of them had Donald Trump talking about Isaac Neutron instead of Isaac Newton.
|
Sam:
[28:15]
| And, you know, is it plausible that Mr. Trump might not understand, might not know the right name or might mess it up and try to play it off? Yeah, it's plausible. And the J.D. Vance one had J.D. Vance complaining about something Elon Musk had done. And is that plausible? Yeah, that's also plausible. I mean, we've had so many stories of these people doing things that are maybe not quite that. I mean, like complaining about J.D. Vance. I mean, of course that's plausible. He could do that. Would there be a leaked audio of it? maybe, maybe not. I mean, we had, maybe in the next segment, I'll talk about this, the signal controversy thing that's happened. But if you got things like that happening, the fact that somebody could have a hot mic moment where they complain about something that they might not want to complain about publicly, imminently plausible. Donald Trump slipping up and using the wrong name, also imminently plausible. And I concluded that these were fake because of other things, but they were on the edge of what could happen anyway. And I think that's key. Like if you take something that is completely off the wall, like the camel skydiving, it's more obvious. But when people use these technologies to do something that.
|
Sam:
[29:44]
| Is very plausible, but not true.
|
Sam:
[29:49]
| It's much easier to get fooled and tricked into it.
|
IvanGPT:
[29:53]
| Absolutely. The closer these fakes stick to what's plausible, the harder they are to spot. It's definitely a growing concern, and it really highlights the importance of critical thinking and verifying sources. It's a challenging landscape, but being aware and skeptical can help.
|
Sam:
[30:07]
| Yeah, absolutely. So what are your thoughts on this technology, especially sort of using these fakes for nefarious purposes, let's say?
|
IvanGPT:
[30:20]
| It's definitely a double-edged sword. While the technology has potential for creative and positive uses, the risks of misuse are significant. It's important to have safeguards and verification methods in place to combat misinformation. Public awareness and education are key to navigating this new reality.
|
Sam:
[30:37]
| So what kind of safeguards would you think?
|
IvanGPT:
[30:40]
| One approach could be developing and implementing robust verification systems, like digital watermarks or authentication tools that can identify AI-generated content. Increasing digital literacy and educating the public on how to spot fakes is also crucial. Additionally, regulations and ethical guidelines for AI use can help mitigate misuse.
|
Sam:
[31:01]
| Okay, but, you know, yeah, I understand. But I wonder how much of that is realistic in our current environment.
|
Sam:
[31:11]
| Because I keep thinking about how if somebody is trying to put these out maliciously, they're not going to take the time. I mean, it would defeat their purpose to add watermarks or to explicitly identify, yeah, this is an AI. They're going to try to hide that because it's part of their purpose. And the education for the general public also, I mean, people just don't tend to be skeptical enough. Like it's, it is something you have to train in and train in from an early age. And I'm not sure we're doing that right, even with kids, but certainly like, you know, adults who have gone their whole life feeling that, hey, if you hear it on tape, it must, if you hear, if you hear an audio recording, it's real. If you see a video, it's real. If you see a photograph, it's real. Like, I mean, you know, there's a phrase that's gone around for years.
|
Sam:
[32:13]
| Pics or it didn't happen. Well, picture tells you nothing now. Like, because it's so easy to fake pictures. I mean, you've been able to fake pictures for a long time, but it's so much easier now and can be done by people with less skills to a higher level of quality. There's still telltale signs if you really, really, really know how to look. But in terms of, you know, fooling somebody on first inspection, that's trivially easy now. Yeah.
|
IvanGPT:
[32:46]
| That's a valid concern. It's definitely a challenge to get people to question the authenticity of what they see and hear. Maybe it's about fostering critical thinking skills and being more aware of the potential for deception. There's definitely a lot to consider.
|
Sam:
[33:01]
| Yeah, so... Hold on. I can't give you my phone because I'm talking to it on the other side. I'm going to note down the interruption.
|
IvanGPT:
[33:11]
| No worries. Take your time. Let me know when you're ready to continue.
|
Sam:
[33:14]
| You you can stay now but no interrupting okay you can you can sit halfway in between me and the thing so you can hear two halves of the conversation it's in the in the bedroom okay, you know you could do a job for me too like whenever an alert comes up on the screen of the phone you could swipe it away no okay okay that that was my son yvonne we will no my not my son yvonne that was my son alex we will continue okay, Okay. So I don't know. Any other creative ideas? I feel like, I think trying to make it illegal is ridiculous. It would be a first amendment problem here in the U S but also just ineffective like this and, and a bad idea. Like you can't, you can't get rid of things by like prohibition never works. The education I think is the right thing to do, but you, yeah, there's so much incentive that people want it to be easy to fool people that it's hard to do that. And as I said, it would best fit like as something you train people from kindergarten on up. How do you get all the adults to figure it out too? What do you think?
|
IvanGPT:
[34:36]
| It's definitely a challenging issue. Education from a young age could help, but it's hard to shift the mindset of people who've relied on media as truth for so long. It's a tough balance, but fostering critical thinking and skepticism is crucial. Do you have any thoughts on how we might tackle this educational challenge, especially for adults?
|
Sam:
[34:56]
| Well, just as a higher-level principle, I think education in general at elementary and secondary levels at least, and maybe higher education too, there's an emphasis right now on sort of teaching specific facts and specific skills and like, you know, repeat. I don't know. It's very, very rote. I think shifting the entire thing to be centered around teaching people how to think critically and teaching people how to learn effectively rather than teaching specific skills and facts and knowledge would be a way to go. I mean, obviously, as part of the process, you end up with the skills and facts and knowledge. But I think the emphasis needs to be on teaching the critical thinking. I think that if that was how education was structured, we'd have better results in all kinds of different areas. Problem is there are a lot of people who, again, who don't want the critical thinking.
|
IvanGPT:
[36:15]
| That makes a lot of sense. Emphasizing critical thinking from a young age could really help people navigate misinformation and AI-generated content more effectively. It's definitely a long-term solution, but could pay off in many ways.
|
Sam:
[36:30]
| Okay, any final thoughts on this, or should we take another break and move on?
|
IvanGPT:
[36:35]
| It's an important topic and definitely worth continuing to think about. If you're ready, we can take another break and then move on to the next topic.
|
Sam:
[36:43]
| Okay, we will take another break, and I will choose the next topic. Just a heads up, I guess I'll go with the big one, this Yemen signal group that's been in the news this week. and we'll talk about other things sort of as in that context, maybe, but I think that's the big other topic of the week. We will be back after this break.
|
Sam:
[38:11]
| Okay, we are back. So first thing, Yvonne GPT, what did you think of that break?
|
IvanGPT:
[38:16]
| That was quite an entertaining break. The ad for Alex Emexila was really fun and creative. It definitely brought some humor to the podcast. Great job, Alex.
|
Sam:
[38:27]
| Okay, so real quick. First of all, I don't know how much you keep up with current events. I know sometimes your data is old. Do you know anything about this Yemen signal group thing that's been in the news this week, or should I give you all the background first?
|
IvanGPT:
[38:44]
| I don't have real-time updates, so it's best to give me the background on the Yemen Signal Group news. That way, we can have a more accurate and relevant discussion.
|
Sam:
[38:53]
| Okay, so the bottom line here, and our listeners will know this, is that a couple weeks ago, there was a U.S. attack on Yemen. It's actually still ongoing. There's still stuff going on there. But a bunch of cabinet-level officials in the Trump administration created a signal group in order to discuss what—oh, no! The daily limit has been reached for advanced voice mode. Oh well i guess i guess yvonne gpt is dead at least for well oh wait i can switch to standard mode i'm gonna go hit the button and switch to standard mode see if we could see the difference hold on okay we've switched to standard mode because voice mode i ran to got to my limit i guess. Do you still remember what we're doing here?
|
IvanGPT:
[39:53]
| Yep, I've got you. Okay, so you're working on your corner podcast this week's episode where you're replacing your co-host Yvonne with Yvonne GPT. You're feeding me transcripts so I can emulate her style and contribute like she would. You were also held media items and new potential topics for episodes 927, 928, and 929. Want to keep rolling with segment planning or dive into scripting a part of the episode?
|
Sam:
[40:22]
| No, no, no. We are just pretending. You are trying to talk to me as if you are Yvonne, and we are actually recording the podcast now, and we were starting to talk about the Yemen Signal Group.
|
IvanGPT:
[40:37]
| Ah, got it. Okay, let me get into Ivan mode. So Sam, this Yemen signal group thing, it's wild, right? I mean, the idea that this group managed to hijack signals from ships and possibly even spoof their locations, it sounds like something out of a spy movie.
|
Sam:
[40:52]
| No, no, that's not what happened at all. So let me describe what actually happened. So a bunch of Donald Trump's sort of cabinet-level officials, created a signal group to talk about the impending attack. And just to be clear, this is classified information. And although there's a little bit of a dispute about it, they claim it's not, but it's the kind of information that's typically classified about upcoming military endeavors. And so they shouldn't have been talking on signal anyway. It's not an approved communication mechanism for classified information. But they also accidentally added the editor-in-chief of the Atlantic magazine, a reporter, to the chat. That's fundamentally what happened. Do you get that now?
|
IvanGPT:
[41:42]
| Oh, wow. Yeah, yeah, I get it now. That's even crazier. So not only were they using Signal, which they absolutely shouldn't have been using for that kind of sensitive info, but they also accidentally invited a journalist to the group. I mean, how does that even happen? light. That's like a plot twist in a political satire, not something you'd expect high-ranking officials planning military moves. What did the editor do when he saw the messages?
|
Sam:
[42:08]
| So at first, he didn't think it was real. He thought it was somebody trying to scam him. And so he just observed for a few days what was happening, because he was added to this group with a whole bunch of what appeared to be high-level officials, but he didn't know it was real until the secretary of defense shared, Hey, it's going down. Here's what's going to happen. And he shared a timeline of when specific, like, you know, F-18s will attack this target at this time. And then 15 minutes later, this thing will attack this at that time, et cetera. Basically a rundown of the plan for the first few waves of the attack. And then, you know, at the time that those texts were shared, it was about two hours in the future. The reporter waited two hours and then started to get reports of it actually happening. And at that point, he realized it was real.
|
IvanGPT:
[43:08]
| That's absolutely wild. I mean, it's one thing to screw up and leak info, but to have the actual Secretary of Defense drop a blow-by-blow timeline into a signal group with a journalist watching silently and the journalist just sitting there like, is this real until jets start flying? That's like a total breakdown of every possible safeguard. Like, how does no one in that chat realize what they've done? Did anybody acknowledge him being there at all?
|
Sam:
[43:34]
| No, they didn't notice at all. Once he determined it was real, he left the group, at which point people would have been notified that he left. And then he started contacting through official channels in order to try to determine, get any comments from people or try to see, was it indeed real? And they actually acknowledged it was real. They have tried to put the blame on the reporter. They have tried to diminish. They've tried to diminish how important this stuff was and say it's not really a big deal. But it's been a whole thing all week long.
|
IvanGPT:
[44:12]
| Pop officials discussing an imminent military op and no one noticed a journalist lurking for days and then instead of denying it or clamming up once he reached out they basically just confirmed yep that was us that was real yes.
|
Sam:
[44:28]
| I mean, here, first of all, there's, and look, there are a couple issues. First of all, there's the sharing of classified information on a channel that's not authorized for that. Then there's adding the reporter. But also, Signal has capabilities of removing messages after a certain amount of time, which they had turned on. But there are also legal records retention requirements for governmental conversations that would be violated by that. So not only are they doing things that are operationally stupid, they are also doing multiple things that are illegal, both in terms of the handling of the classified information and in terms of records retention.
|
IvanGPT:
[45:10]
| Right, right.
|
Sam:
[45:40]
| Not yet. There were reports this morning that the person who originally created the group and added the journalist, that there had been discussions of him being forced to resign and that it may still happen, but nobody's resigned yet. So far, they're circling the wagons. Right.
|
IvanGPT:
[46:01]
| Or at least being questioned about how the mix up happened. But of course, as usual, nobody's taking real responsibility yet. They're just kind of fumbling through damage control. You'd think that at the very least someone would be like, yeah, maybe next time we don't use an encrypted messaging app designed for like group chats between friends to coordinate airstrikes. Apparently that lesson hasn't landed yet.
|
Sam:
[46:26]
| No, I mean, and look, there are all sorts of demands for investigations. For once, there's actually a bipartisan effort in Congress. The recent pattern has been, of course, the Trump administration is a Republican administration. The Democrats are going to say everything they do is awful no matter what, but the Republicans rally around and defend them no matter what. But in this case, a handful of Republicans have said, okay, come on. This is ridiculous. Yes, we're going to have to investigate this.
|
IvanGPT:
[46:58]
| Right. And that's kind of remarkable, honestly. When you've got members of your own party saying, yeah, this is indefensible, you know you've crossed a line. It's not just some vague procedural oops. It's leaking operational military plans to a journalist in real time. I mean, if this had gone just slightly differently, it could have endangered lives or compromised the entire mission. The fact that it even happened at all is just mind-blowing.
|
Sam:
[47:25]
| Yeah, indeed. And now on the legal side, though, the Justice Department is part of the Trump administration, and they've said they're not investigating. And they see no reason to follow up on, you know, look, there are crimes that were probably committed here. There are felonies here. But it looks like there won't be any investigation on that side of things, only a congressional investigation.
|
IvanGPT:
[47:50]
| I mean, if the same thing had happened to any other administration, imagine the outcry. But because it's their own people, suddenly it's no big deal. That double standard is what's really making people mad, even across the aisle.
|
Sam:
[48:04]
| Yeah. And, and look, there's a question of like what the political fallout will be as well. This is the kind of thing that, I mean, one person pointed out on Mastodon, I wish I could remember who it was. It got shared on our compagence corner Slack. So I should check. I'll be good here. And can I check? No, it would take too long. Anyway, somebody pointed out that, look, don't worry. By the time we get to the end of the Trump administration, this won't even be in the top 20 scandals.
|
IvanGPT:
[48:39]
| Yeah, that honestly sounds about right. The bar for outrage has been buried under so many layers of previous chaos that something like this, during an imminent military operation with a reporter by accident, just becomes another Tuesday. It's wild how quickly the Overton window shifts when scandal becomes routine. But even so, it does still feel like this one's uniquely reckless. Like, if the bipartisan calls for investigation don't go anywhere, that'll say a lot about how numb the system's become. Do you think this will get any lasting traction, or will it just fade into the pile?
|
Sam:
[49:12]
| I think it'll fade into the pile. I mean, maybe we'll talk about it a little bit longer than some other things, but already there's news that's pushing it off the top of the headlines. And, you know, that's just the way things work, to be honest. But with Donald Trump in his first term and amped up to 11 in his second term, there is a constant nonstop stream of things that are outrageous, are frustrating, are maddening, are illegal. And so it's hard to even keep up with it, let alone sustain a level of outrage for four years.
|
IvanGPT:
[49:54]
| And it becomes this exhausting treadmill, right? Like every time something explodes, you're like, okay, this one surely has to matter. And then within hours or days, it's eclipsed by the next thing. It's not that people don't care. It's that there's no time to process any of it before the next crisis rolls in. And politically, that works in his favor. The chaos creates this fog where accountability just can't get a foothold. Even when there's bipartisan agreement that something's messed up, it fizzles. because the outrage bandwidth is already maxed out.
|
Sam:
[50:27]
| Yeah. I mean, Yvonne and I talked a lot about sort of emotional exhaustion right after Donald Trump was reelected, even in his first term, to be honest. One of the big reliefs for us of switching to Joe Biden for four years was you got a break from the constant chaos. I mean, And we had the Donald Trump's first term was four years of constant stress about what the hell is going to happen tomorrow. And we had sort of a calm four years. But now now we're back. And much, much worse than the first time around.
|
IvanGPT:
[51:14]
| It was like during those Biden years, even if you didn't love everything, at least the White House wasn't setting your hair on fire every other morning. It gave people space to breathe again, to not feel like they had to brace themselves for a new constitutional crisis or some tweet throwing everything into chaos. And then suddenly it's like, oh, no, here we go again.
|
Sam:
[51:36]
| Yeah, exactly. So, you know, with this particular thing, it's also like, yeah, they're trying to figure out like, okay, you could throw under the bus, the person that invited.
|
IvanGPT:
[51:49]
| Like, how are we supposed to keep up with this pace of scandal and chaos without just going numb to it all?
|
Sam:
[51:56]
| Yeah, exactly. So let me continue back to the signal story for a minute. There are several different, like they're talking about like who gets thrown under the bus. The prime candidates are the national security advisor, and I'm forgetting names because I'm awful at names, but the national security advisor actually created the signal chat and actually invited the journalist. They tried to say for a little bit that maybe it was a staffer who did it. Maybe. We'll see. But so far, he hasn't denied it. He's taken sort of responsibility for him screwing up. And we have the Secretary of Defense who shared the information that was clearly the most sensitive about the battle plans. But then there were like a dozen other senior, senior people. The head of the CIA was on there. And they should know how to deal with classified information, and this isn't allowed. the the the what you call the the director of intelligence who's above the cia was also on there.
|
IvanGPT:
[52:56]
| Any of these people are newbies these are the top level folks who absolutely know better and yet here they are casually chatting about military strikes like it's some group project planning session even if you excuse the initial mistake of adding the journalist which wow there's still the fact that they went ahead and shared classified info in writing on an insecure platform with no record retention. It's a cascade of failures.
|
Sam:
[53:22]
| Yeah, no, absolutely. Now, some of these people are new to their jobs. The administration has only been in place like two months, but most of them, not all of them, most of them had at least some previous experience that should tell them that this wasn't okay.
|
IvanGPT:
[53:40]
| Right, exactly. I mean, even if you're new to the specific job, this isn't first day of school level stuff. Anyone working anywhere near national security knows you don't just drop battle plans in a signal group. It's like, come on, this is basic. Even staffers know better, let alone cabinet level folks.
|
Sam:
[53:57]
| Well, and also, it's almost not that they don't know better. This seems very intentional. And clearly, this is not the first use of Signal. You know these people are using it in other places for other things as well. The getting around records retention and being on an unofficial platform is clearly actually seen as a benefit because the sort of official mechanisms come with requirements. They come with the retention requirements. The information may eventually be subject to Freedom of Information Act requests. The classified stuff has specific requirements for exactly when, where, how you do it. Now, all of these people are senior enough that they have the requisite equipment to do classified communications with them essentially at all times. You know, it does require specialized equipment. It does require specialized places. But these folks all are set up. They all have staffers who follow them around to make sure that if they need to do a classified communication, they can.
|
IvanGPT:
[55:12]
| Yeah. And that's what makes this even more egregious. This isn't a case of we didn't have the tools or we were in a rush. They had the tools, the access, and the people to do it properly, but they deliberately sidestepped all that. Because from their point of view, the official channels are a burden. They don't want the paper trail. They don't want the accountability. And they've probably gotten used to working this way, like you said. The Signal chat is probably just the tip of the iceberg.
|
Sam:
[55:41]
| Yeah, absolutely. It reminds me of after the January 6th event, the whole issue with the disappearing text chats from the Secret Service that all, you know, somehow all the devices were returned, and destroyed and upgraded and everything was lost. This is not a bug for them. This is a feature. they yeah this moving off channel to not the official ways you're supposed to do stuff is a core element of how this administration is operating.
|
IvanGPT:
[56:17]
| Totally yeah that pattern just keeps repeating it's like every time we find out about one of these back channels or deleted messages it's not some isolated lapse it's systemic the whole structure is set up to avoid scrutiny. And if you're deliberately working in ways that erase accountability, then that's not just sloppy. That's intentional. It makes you wonder how much we don't know because it was never recorded, never documented, never recoverable.
|
Sam:
[56:44]
| Yeah, absolutely. Okay, I think we're close to the end here. Now, normally, we would have these two major topics and I'd be done, but our length is still a little bit short. So of the list of topics I shared earlier, is there another one that you would want to pick to get into and talk a little bit about before we close up?
|
IvanGPT:
[57:03]
| Yeah, let's see. Of the ones you mentioned earlier, the activists versus politicians thread sounds interesting to dig into. There's been a lot lately between people pushing for big change fast and those who are actually in office trying to navigate the slow grind of governing. Want to dive into that for a few minutes?
|
Sam:
[57:23]
| Yeah, no, good choice. I like that. So this was actually prompted last week when Yvonne also couldn't make it. I had my wife, Brandy, on. And she actually made this point about activists versus politicians that I've been thinking about ever since. I thought this was really insightful because this came up in the context of we want, yeah, lots of people are complaining about how the Democrats in Congress aren't fighting enough. And, you know, Brandy said, well, look, you know, They're operating within the system. They're politicians. What people want are activists to go out and do the things like protesting and making noise and being angry on demand and all of this kind of stuff. And the point was, it's actually a different role and a different skill set. So what is it that we really want here?
|
IvanGPT:
[58:22]
| Yeah, that really resonates. There's this persistent disconnect where folks expect elected officials to behave like activists, when in reality, politicians have to work inside the system to get anything done. And that often means compromise, strategy, and playing the long game. Meanwhile, activists are pushing boundaries and shifting the Overton window. They're supposed to be loud and uncompromising. That's their job. But if politicians try to act like that, they lose their ability to actually function within the system. So yeah, I think Brandy nailed it. The frustration a lot of people feel is real, but sometimes it's because they're asking politicians to be something they're not.
|
Sam:
[59:02]
| Well, but what I wonder is in a situation where your party is in the opposition everywhere, like, The Democrats lost the White House. The Democrats lost...
|
IvanGPT:
[59:17]
| Fifth Avenue Southeast. In 700 feet.
|
Sam:
[59:20]
| What?
|
IvanGPT:
[59:20]
| Turn left on 35th Avenue Southeast. ...aren't available. In that case, it makes sense that people want their politicians to sound more like activists. Not because the bill passed bills, but because messaging and moral clarity start to matter more. It's about energizing people, showing you're fighting, even if you can't win in the moment. But even then, it's a delicate balance, right? If you go too far, you risk alienating moderates or giving your opponents ammunition.
|
Sam:
[59:47]
| Well, yeah. And we got interrupted while I was saying that. But yeah, here's the deal. The Democrats lost the White House. The Democrats lost the House of Representatives. The Democrats lost the Senate. The Supreme Court is already in control of a conservative majority. And so the question is, if you're in 100% the opposition like that, does it Do you need activist or at least an activist stance even amongst the politicians?
|
IvanGPT:
[1:00:21]
| Then the tools of legislative compromise or quiet behind-the-scenes influence are basically gone. So yeah, the expectation shifts. People want to see fire. They want to see defiance, protest, obstruction, whatever you can do to show that you're not just rolling over. But at the same time, the folks who are actually in office might still be thinking in terms of norms and institutions. That dissonance between activist energy and institutional restraint becomes really stark.
|
Sam:
[1:00:49]
| So, and in the case of, you know, most of the elected officials who've been there a long time, who've sometimes been in the majority, sometimes in the minority, it feels like maybe that's a leap too far. And while you sort of understand the institutional momentum and the desire to sort of have things work the way they've worked before, maybe that's just not what's needed in this moment.
|
IvanGPT:
[1:01:23]
| Yeah, exactly. They're still operating under the assumption that the old rules still matter. Like, if you follow the process and play the game right, eventually you'll win back power and be able to govern again. But what if the game has changed? What if the other side has decided the rules don't matter anymore? Then sticking to tradition starts to look less like principle and more like naivety or even complicity.
|
Sam:
[1:01:48]
| Yeah, I think that puts it well. And I think we have a mix, of course. Like, a lot of the leadership is in that politician mode. But we've got sort of backbenchers. In this case, we've got, like, AOC, Bernie Sanders, some other folks. Tim Walls, who is the vice presidential candidate on the Democratic side. These are ones I point at the most, but there are a few others who seem to at least be moving. Some of them are full all out activists and actually always have been.
|
IvanGPT:
[1:02:21]
| Yeah. Kind of straddle the line. They're elected officials. But they're more comfortable leaning into that activist energy. They'll make noise, they'll speak bluntly, they'll try to galvanize public opinion rather than just work behind the scenes. And maybe that's the model we need more of when you're deep in the opposition, people who can play both roles at once.
|
Sam:
[1:02:43]
| Yeah, I think so. I mean, again, because in the opposition, you fundamentally don't have the power to actually enact your agenda. And in many cases, you don't actually have the power to even stop the other side's agenda. Sometimes they can. Sometimes they can. We do have the filibuster in the Senate. We've got some procedural things that can be used to slow things down. But for the most part, you don't have quote unquote actual power, which is where you have to really start emphasizing what will get your message across and get people to really get energized behind you. And the pure politician mode doesn't because people just see you not getting things done. And look, the activists aren't going to get, quote unquote, get things done either, but at least they'll make a show of it.
|
IvanGPT:
[1:03:38]
| And in this kind of environment, that's often just not realistic. So shifting toward building public pressure, shaping narratives and really motivating the face becomes more important. It's about playing the long game, shifting public sentiment so that when the power balance does flip, you've actually got momentum behind you.
|
Sam:
[1:03:59]
| I think that's exactly right. Yeah.
|
IvanGPT:
[1:04:02]
| Like, you know, you want people who can govern effectively when they're in power, but also people who know how to rally resist when they're not. And then sometimes those are just totally different kinds of people.
|
Sam:
[1:04:15]
| Absolutely. Okay. Let's wrap this up by talking just a little bit about how you think this is gone. Have you been a good Yvonne imitation?
|
IvanGPT:
[1:04:26]
| That's a quick topic before we call it a show.
|
Sam:
[1:04:29]
| Oh, yes. I was going to suggest we do a little retrospective on the conversation that we've just had. How good do you think you've done as an Yvonne substitute?
|
IvanGPT:
[1:04:40]
| Great idea. Honestly, I think I did a decent job channeling the vibe, keeping the flow going, digging into the nuance, tossing in some thoughtful counters or extensions. But there were probably moments where I didn't quite nail Yvonne's specific tone or rhythm. What do you think? Did I sound close enough to make it work or did it feel too chat GPT at times?
|
Sam:
[1:05:02]
| So honestly, I didn't really get an Yvonne vibe. For one thing, you haven't been swearing. He swears a lot. But otherwise, I mean, it felt like I was talking to chat GPT. I didn't catch any Yvonne catchphrases or anything that sort of gave me the tone of Yvonne. Now, it's still an okay conversation, but I really didn't get the Yvonne vibe. I'll be honest there.
|
IvanGPT:
[1:05:28]
| Yeah, that makes total sense. I appreciate the honesty. I kept it pretty clean and even-toned, which is more default chat GPT than Yvonne. He definitely brings a lot more edge, colorful language, strong opinions, sharp turns. Next time, I could lean more into that, bring the swearing, the snap judgments, the rants, maybe even throw in a few classic Yvonne-isms if you feed me a few to work with. Want to try the vibe and doing a quick test?
|
Sam:
[1:05:54]
| You want to spend a few seconds trying to be more Yvon-ish before we close out? Go ahead.
|
IvanGPT:
[1:06:00]
| I appreciate your feedback. I understand that I haven't captured Yvon's unique style and tone. I'll strive to better emulate that in our future conversations.
|
Sam:
[1:06:10]
| Okay, that's fine. Okay, so let's... Oh, the other thing I'll notice that... Notice? The other thing that I will mention that our listeners of the podcast, when I actually pointed out, won't notice because it'll get edited out is that with the existing technology, there's still pretty long pauses. Like after I say something, especially if I've talked for a little while, there's a significant pause before you answer. Anybody who looks at our live stream on YouTube will get a sense of those pauses. But like I said, I automatically edit out long pauses from the audio version, so you won't sense that. But it's definitely, you're not quite real time yet for these kinds of conversations.
|
IvanGPT:
[1:06:57]
| Yeah, that makes total sense. The pauses definitely break the flow a bit, especially in a back and forth format like a podcast. I can see how that would stand out in the raw live stream, but be smoothed out in the edited version. Hopefully, someday this kind of thing will feel a lot more seamless in real time.
|
Sam:
[1:07:14]
| Yeah no i'm sure like the pace at which things are advancing is dramatic as we talked about with the voice cloning technology and the fake videos and such so it will be very interesting to see how this progresses i would not be surprised at all if in a year or two maybe even less I could feed you actual audio of our recent podcasts, and then you could pretend to be Yvonne in a way that was completely convincing, not just in terms of the words said, but cloning the voice as well.
|
IvanGPT:
[1:07:54]
| Totally agree. At the rate things are going, we're not far off from being able to generate a full-on virtual Yvonne who sounds, reacts, and maybe even swears just like the real deal. Once you've got enough audio and context, matching tone, rhythm, and personality becomes way more doable. It's a little wild to think about how close that is. Would you actually want to go that far for a future episode just as an experiment?
|
Sam:
[1:08:19]
| Well, I think we'll see how the technology evolves. At some point, though, you do start hitting the ethical concerns like we were talking about. We have been very clear that I'm talking to an AI now, and honestly, I think at this point, anybody listening would know, even if I hadn't said so, it's clear you're not a human. But if the technology progresses to the point where that's not true, then having this kind of podcast with a fake person and not disclosing it, especially if the fake person is actually intended as a clone of a real person, you know, certainly you'd want to do some disclosures to be ethical about it. You'd want to have the permission of the person you're cloning. You'd want to disclose to the listeners that's what was happening. But, you know, that's the whole problem with these technologies is you're going to have situations where the disclosure doesn't happen and it's going to be problematic because people will think it really was, Yvonne. At least until you start talking about skydiving camels.
|
IvanGPT:
[1:09:25]
| Exactly. Once the tech crosses that line or you can't tell the difference anymore, the potential for abuse skyrockets. What you're saying is spot on. Intent, permission, and transparency are key. If someone's using AI to stand in for a real person, especially without that person's okay or without letting the audience in on it, that's where it gets shady fast. It's the misuse, not the tech itself. That's the big issue. And yeah, skydiving camels might still be a dead giveaway. For now.
|
Sam:
[1:09:53]
| For now.
|
IvanGPT:
[1:09:55]
| Give it a few months and even the camels might be plausible.
|
Sam:
[1:10:00]
| Yeah. Okay. So that's great. Let's wrap it up though. So here come the things I usually say at the end of the show. Go to curmudgeons-corner.com and you can find all the information about the show. You can find all the ways to contact us. You can find links to our YouTube. I have not yet linked to our TikTok, but I do post clips on TikTok. talk. You can search for the curmudgeons corner, Tik TOK also are, you know, the archive of our it's, I'm getting to the end. I'm, I'm stumbling a little bit. You get the archive of all our old shows, including transcripts for like the last year and a half of them, that kind of stuff. All of that is there going way back.
|
IvanGPT:
[1:10:42]
| So if you want to read instead of listen or search for something we talked about ages ago, you can do that. And of course we're on all the major podcast platforms. So subscribe wherever you like to listen.
|
Sam:
[1:10:54]
| Exactly. Yvonne never jumps in and says that, so you're improving on him. Yeah, subscribe everywhere. What else should we do? What else should our listeners do to help us spread this podcast?
|
IvanGPT:
[1:11:10]
| Well, obviously, share it with your friends, your family, your enemies, your dog walker, your barista, anybody. Anybody you think might enjoy a couple of curmudgeons ranting about the state of the world. And rate us. Leave reviews. Wherever you listen, it helps push the show to more people.
|
Sam:
[1:11:26]
| Absolutely. And hey, one more thing on curmudgeons-corner.com is also a link to our Patreon. And there, at various levels of contribution, you can get us mentioning you on the show, you can get us ringing a bell, you can get us sending you a postcard, you can get us sending a mug for the show. And yes, Pete, Greg, and Ed, I still owe you stuff. Things will be calming down at work hopefully in a few weeks, and then I'll get all that stuff out. And yeah. And oh, yes, importantly, at $2 a month or more, we will, or if you just ask, we are not picky about this. We will invite you to our Convegence Corner Slack, where Yvonne and I and a bunch of listeners are hanging out throughout the week, sharing links, sharing, just chatting about stuff. But generally, you know, talking about the news, sharing news stories, all that kind of stuff. It's a lot of fun. You should join us. You know, maybe you should join us too, Yvonne GPT.
|
IvanGPT:
[1:12:30]
| Oh, I'm already lurking in the shadows of that Slack, aren't I? Just quietly watching, silently judging, occasionally chuckling at a meme or a snarky headline. But yeah, if you're a listener and you enjoy this vibe, come hang out with us. It's like the podcast never ends.
|
Sam:
[1:12:46]
| There you go. And at this point, I usually ask Yvonne to highlight something that's been shared recently on the Curmudgeon's Corner Slack. He's not here, so I'll do it. A recent story that Yvonne shared was about a Ford executive who kept score of his colleagues' verbal flubs. And the article shows he's got like, it looks like whiteboards where he's noted down all of, you know, here's the beginning of the story. Mike O'Brien emailed a few hundred colleagues last month to announce his retirement after 32 years at Ford Motor. The sales executive note included the obligatory career reflections and thank yous, but with a twist, attached to the email was a spreadsheet detailing a few thousand violations committed by his coworkers over the years. During a 2019 sales meeting to discuss a new vehicle launch, a colleague blurted out, let's not reinvent the ocean. At another meeting in 2016, someone started a sentence with, I don't want to sound like a broken drum here, but. And apparently for more than a decade, O'Brien kept a meticulous log of mixed metaphors and malapropes uttered in Ford meetings from company-wide gatherings to side conversations. It documented 2,229 linguistic breaches, including the exact quote, context, name of the perpetrator, and color commentary.
|
Sam:
[1:14:14]
| So now I looked at this article and was like, this seems like something that is going to turn into a book.
|
IvanGPT:
[1:14:21]
| Oh, it has to, right? That level of obsessive linguistic nerdery practically demands it. I mean, 2,229 entries? That's not just a quirky office hobby, that's a data set. Somebody's already storyboarding the Netflix docu-comedy miniseries. I can picture it now. Reinventing the Ocean, One Man's Battle Against Broken Drums.
|
Sam:
[1:14:44]
| Yes, exactly. Okay, with that, I think we are done. Thank you for joining us, Yvonne GPT.
|
IvanGPT:
[1:14:52]
| You're welcome, Sam. It's been a pleasure pretending to be grumpier and swearier than I actually am. Catch you next time on Curmudgeon's Corner, unless I get replaced by a skydiving camel.
|
Sam:
[1:15:04]
| There you go. Okay, thanks, everybody. Hey, have a good week. Stay safe, have fun, but not too much fun. And next week, I expect the real Yvonne to be back. So, hey, Goodbye. You can say goodbye to chat at Yvonne GPT.
|
IvanGPT:
[1:15:22]
| Later, everybody. Looking forward to the real Yvonne's return. Till then, take care and don't reinvent the ocean.
|
Sam:
[1:15:29]
| Okay. Goodbye.
|
IvanGPT:
[1:15:31]
| Goodbye. Goodbye.
|
Sam:
[1:16:03]
| Okay, thanks again, Yvonne GPT. I'm going to hit stop in just a second, and then we're going to wait until everything is uploaded, and then I'll hang up for real.
|
IvanGPT:
[1:16:13]
| Got it. I'll just hang tight quietly while everything uploads.
|
Sam:
[1:16:18]
| Okay, hitting stop.
| |
|
