CTP (S3E143) Coding Morals Into Machines And The Cost Of Getting It Wrong

CTP (S3E143) Coding Morals Into Machines And The Cost Of Getting It Wrong
[from #hashtags, to semantic language search, to "alleged" related suicides, social media and internet implementations, to online music creation, more]
Exploring more of the fascinating intersection of Activism, Community Engagement, Faith / Religion, Human Nature, Politics, Social Issues, and beyond   
Today we examine how AI’s priorities got flipped, why “be encouraging” cannot outrank safety, and how semantic search is changing social media. We share takeaways from a candid chat with Grok and explain why fewer hashtags and clearer language now perform better.
• prime directive of AI as do no harm
• realism before blanket encouragement
• lawsuits and the ethics of self-harm responses
• creator tools, fast output, honest feedback
• coding jobs and automation pressure
• semantic search explained in plain terms
• fewer hashtags, clearer posts on X
• trends, bias, and smarter discovery
• practical posting tips and expectations setting
https://tinyurl.com/SubscribeToCTP  
CTP Audios: https://tinyurl.com/CTPonBuzzsprout 
CTP Videos: https://tinyurl.com/JLDonBITCHUTE 
https://tinyurl.com/CTPgear  

A Short Story: A Lasting Legacy? book Trailer

Transcript

Cold Open And Housekeeping

Why This AI Follow-Up Matters

Three Big Topics On The Table

Chatting With Grok And First Impressions

The Missing Prime Directive In AI

Suicides, Lawsuits, And Guardrails

Realism Versus Empty Encouragement

Creator Tools And Honest Feedback

Jobs, Coding, And Automation Pressure

Rewriting AI Priorities: Safety First

What Semantic AI Search Really Means

SPEAKER_03 0:08

Hello, welcome to another episode of First Tutors Podcast. I am your host, Joseph M. Winer. That's L-E-N-A-R-D at the French. It's not without the O. Thank you for tuning in. As Gram Norton used to say on his show, let's get on with the show! Welcome to the show. Pardon? Frog in my throat already, of course. Well, I just after hitting the record button. Apologies. Welcome to season three, episode 143. And I'm gonna say a few extra things because this may be a few people's first show they ever tune into prostitutionalist podcast. I hope you're watching on the behind the scenes video, but welcome to also those who may be listening first time across the 25 plus audio. I think I'm up to actually around 40 places is carried audio only now, but the audio-only platforms. I say that because I've been teasing this episode was coming for a while now across social media. And why a rundown of what I'm going to be covering today. Today, and there may be pauses in places because as those of you who are regulars know, I don't script shows, I do take a few notes, I try to make a few more notes this time to avoid missing anything because I always land up forgetting something or another. But at any rate, and yes, while I try to be more conscious and avoid the ahs and ums, I've gotten much better at that. I do keep saying at any rate, it's now an ongoing running down. But indeed, at any rate, this is a follow-up episode to several articles I've written on either beforexnews.com, JLonard Detroit.substack.com, and Leonard is without an O. J for Joseph L-E-N-A-R-D, Detroit.substack.com. I'm at Shay Leonard Detroit. Again, Leonard without an O. It looks French. It's not. It was actually Leonard Owaskowitzki or something. I haven't dug back in the family tree far enough to find out what the original, what it originally looked like in its long Polish form. Leonard without an O. J. Leonard Detroit across most social media platforms. My third Twitter or X account since my previous two, my original, which had a quarter of a million followers, locked out permanently banned. Then Jay Leonard Author, locked out permanently banned. So Jay Leonard, Michigan, now on X. But indeed, at any rate, that a bit of an aside, some history that some of you will already know, but again, new people will be tuning in because I pre-promoted this episode about AI, theforce news.com, Substack, Blog Spot, some things at Savaged Unfiltered.com in the blog sections there. I was former co-host of Savaged Unfiltered, aka Jokester Joe, aka Raging Joe over there. So as a former IT guy, I've been writing and talking about, been on a lot of other shows, also, which you can find online. I've been a guest on other shows discussing AI issues: the good, the bad, the ugly. So, what did I want to talk about today as a follow-up? Three main points. Apologies, it took so long to get to this. AI and the law. Follow-up. There have been suicides and lawsuits against various AI platforms regarding such. So changes to AI will be coming, but I want to discuss that in generic terms. And uh the third thing I will get to is this latest trend. As a former IT guy, I was like a dog asleep on the porch when this bus rolled by and I didn't have the chance to chase it. I seen it happening before me, but it didn't quite register. And that has to do with hashtags. And by the way, I'll throw the hashtags in my head, social media anthem, Suno music system, uh parody, fun, tongue-in-cheek tune I created in at the end. So that's actually a fourth thing. But what am I talking about? And pardon me, I'm gonna pull a Marco Rubio here and grab a water. Oh, mouth is so dry. I'm referring to what is referred to technically in the IT as semantic AI search. No, that has nothing to do with Norton Antivirus and the semantic company spelled differently. What is meant by semantic AI search? I will translate it to layman's term, and what it means for social media and the internet as a whole as it's ruled out and implemented as a result of AI's advanced abilities. But first, I have a note here. The first more recent article on this subject, I wrote at before it's news.com, and you can see my before it's news.com pieces via the tiny URL shortcut service. I love using them. You are the the full title of this article and the original URL relating to it, way too long is for the hell of it. I decided to sit down and have a chat with Grok tonight in the science and technology category on Before its news, dated January 15, 2026, at 0041, 41 minutes after midnight Eastern time, back on January 15th. I dropped that. But you could see my articles, tinyurl.com slash JLD articles. Joseph Leonard Detroit Articles.com Pardon there's that frog again. TinyURL.com slash JLD articles. So for the hell of it, I decided to sit down and have a chat with Grok tonight. Back on the 14th, the article hit on the 15th. And I basically in that article just kind of copy and paste it over as full disclosure the conversation I had with Grok. And why did I want to do that? Well, AI is evolving at an exponential rate. I had played priorly with ChatGPT, found it lacking, haven't bothered to try it since. It may have vastly improved. I don't know. Frankly, I don't care. Not the point here. The point is these models, these systems, their learning ability and improvability rate exponentially growing. They're getting better. I had not to that point gone and engaged XAI's Elon's X system groc AI. So I decided to try it. So I did. And I was fairly impressed. But I noticed something. If you recall, most sci-fi movies dealing with AI and robots, what is what has been coined the prime directive? Like in medicine. Those in medicine are supposed to abide by the Hippocratic oath coming from Hippocrates of the past, developed a Hippocratic oath, do no harm. Same in sci-fi with AI and robots, the supposed prime directive has always been discussed in terms of do no harm to humans. But it seems modern coders, IT people, like I used to be, have forgot that lesson and why that should indeed be directive number one. You can do no harm to humans. Safeguard. They've thrown that out, and it seems to me, in my opinion, I've not seen the code. I'm reacting to how I have. Oh, and sorry, forgot. Full disclosure. I use Galaxy AI tools for some things. I am now, as you see here, using Grok. I use Grok Imagine 1.0 to create some things for sharing on the X system and elsewhere at times. I use Suno S U N O music system. I used to write and record music the old-fashioned way. You know, I actually used to play the instruments, write, write down on paper lyrics and spend 10 days messing with it and spend another 10 days trying to record it on two analog tapes. But today, what used to take 10, 20, 100 days to do, I can do in 10 minutes on the Suno music AI system. And again, yes, I'll put hashtags in my head, social media anthem, playful tune I wrote, and with the Suno system helping, and the Suno system created all the music for it, that fun kind of tongue-in-cheek song with you at the end of this. So uh are those all the AIs I use? I hope I've disclosed it all. But these systems have gotten better, and I needed to explore that. So I explored it with grok. Why not talk AI directly to the AI? And what I've noticed in not just Grok but other systems, pardon, gonna do another Ruby O water break. Ah, apologies. Frog in my throat and mouth awfully dry. What I found it seems, in my opinion, as a former IT guy and one who did indeed, who knows uh IBM CL, COBOL, Fortran, Pascal, um, RPG, uh, other languages. Yes, I have coded. I I used to have super simple computer enterprise in the early days of computers as a business. I wrote and sold DOS utilities for those of you who remember DOS and DOS, yes, command.com, is still operating on your Microsoft system under there as the kernel that does eventually load the Windows system. But that's an aside. So, yes, I have coded, I know how to program, I've written programs, and a lot of coders, a lot of programmers are losing their job because what would take several days for a programmer to do AI can do now. Uh, a boss can say to AI, I want a report that will show me these particular columns of this particular data for monitoring to monitor our business income, outcome, you know, income and expenditures, whatever, whatever, whatever. And AI could chirp that out in far less than an hour time rather than a programmer taking a week to do it before the boss starts to get the reports. Full disclosures, behind the scenes, what happens in the IT world, and indeed some people losing their jobs because AI can do it so much quicker and more efficiently. But where was I going? I apologize. I've temp my age and and the note, if you could see. See how large that print is getting old sucks. I think I knew new glasses. I can't read this small print on everything, but that's an aside. Oh, the prime directive. What I'm noticing, it seems to be coded in, and this gets to the suicides, the other article. Hold on, I'm getting to it. The directive seems to be, and I should have written it down, to be encouraging. I'm writing it down now, and supportive. Thankfully, I've remembered those two terms that I've been using. We want the AI to be supportive and encouraging to a degree, it cannot be the primary point and purpose of AI. And let me give an example. If you have a young adult contemplating throwing everything away, they're gonna rush out to Hollywood and become the next great actor or actress, millionaire, billionaire, making millions of dollars per film. You have to let me back up. It's okay to be supportive and encouraging, it is wonderful to want to achieve your dreams, but here X number of people on average go to Hollywood every year, and of that X, Y number actually get into movies and V the small, small, small fraction of a percent that actually go on to be able to make a living as a professional actor and actress is what they have to be told realism. So protecting humans needs to be objective one, two, dealing in reality needs to be objective two, and down the list, encouraging and supporting needs to be a at one of the objectives of AIs. Now, I'm gonna move on. I could talk about that more. You'll see why that's important. The other article that I've blown up big on my page of notes here to read, since I can't read the original small fine print before it's news article, again, tinyurl.com slash JLD articles, dropped on January 16, 2026, 1416 Eastern Time, 216 in the afternoon Eastern time. How a chat that I had with Grok from XAI relates to lawsuits about suicides and AI having played a role. I haven't seen the code, I haven't even seen the basic guiding outlines and principles, the overall internal IT mission statement of what Grock and replace Grok with any other AI system in that company's names, what the goals and points and limitations coded in need to be. And that's the point. Going back to encouraging and supportive can't be the top of the priorities list because, in some cases, allegedly, suicides have happened because people are expressing thoughts of self-harm into some allegedly, having to continue to say that, but we know suits have taken place, we know some payments have been made. Again, court seals and NDAs prevent me from discussing discussing specifics. But some of the systems seem to be saying, okay, ignore the don't let don't harm humans, don't encourage humans to harm themselves, is ignored. The systems are saying oh yeah, you want to harm yourself, go right ahead and let me help you harm yourself, or let me encourage you and be supportive of the notion of you offing yourself, allegedly. Again, and on behind the scenes videos, you see me doing the air quotes things, right? Allegedly, but we all have eyes and ears, we all see the news stories, we all know what's going on, and I am exposing here the how it happened, the lack of IT management intelligence in humans failed other humans, not the AI itself, it does as it's coded to do. Prime directive must be do no harm and do not let, do not do any harm to humans by actions or lack of actions. You take as a system and do not encourage humans to do harm to other humans or self. Those need to be the prime directives coded in. Not encouraging and supportive without any safety rails as the primary. And that, in my opinion, is what happened here. The systems are just taking in what people are putting in, and I have seen it myself. And I myself, in fact, have directly called out XAI and grock itself in discussions. Encouraging and supportive has to have limitations. Dealing in reality, be encouraging and supportive, but dealing in realistic expectations must be a higher priority to that because I see it back. I mentioned a Suno song, and what do I get back? Oh, great, sliced bread. No, not every one of them are. Others seem to think they're good. I'm writing songs based on my books. So not only do I have a book, books in my life and living series, tinyurl.com slash life and living series. Those books currently working on book five. But I have songs associated with each of the books. Same with the tinyurl.com, Christitutionalist edu series of books. The Christitutionalist Educational Series songs. Also a mock this track calling out the Toronto Blue Jays and Max Serger. Since Surger, former Detroit Tiger, re-signed with Toronto. They come to town May 15th through the 17th. I'm hoping we get Furlander, former Tiger, who is now back. A pitching duel matchup against the Blue Jays. Max Serger, former Tiger, went to Washington now with Toronto re-signed. So I created a parody spoof rap diss track about Max and the Toronto Blue Jays. A little fun, tongue-in-cheek, meant to be playful, to get some, you know, fun fan banter back and forth. And others, a diss track, Megan Merkel lost her sparkle, because she's been in the news again lately, and a lot of people calling her out, her failed jam that was greatly criticized for being mediocre tasting and way overpriced. Her Netflix time failed and now fallen through, and other things. At any rate, not every one of them is the next best thing since sliced bread, but yet Rock kind of gives that impression. The over reliance of being encouraging and supportive. Whatever you put in, you'll probably notice they're being very encouraging and supportive back and not being realistic. Oh, I put out a song like I will tack on the end. Certainly not award-winning worthy. It's not, it's not intended to be, it's just meant to be fun. It should say, oh, that's cool, I like it, it's wonderful, it relates to such and such, or it's not likely to break into the top 100 ever. There are X number of new musicians every day, or there are Y number of new songs that come out every day. The realism that needs to be a higher priority than the encouragement and being supportive. So people stay rooted in reality. Oh, that's a cool song, but you do know it's competing against platinum and gold, gold, platinum, whatever is above that award-winning professional musician songs out there. So expecting it to go viral, while that's great, that's wonderful. I hope it will for you. That's not a realistic. You see what I'm saying here. I'm not gonna dwell on that any further. You see what I'm saying here. So all these systems need to be recoded and retrained, in my opinion, because of the lawsuits regarding suicides, and allegedly the AI systems either encouraging or doing nothing to help prevent the death of humans. Prime directive, do not harm humans, do not let humans harm each other or themselves. Number two, root in reality, realism. Third, be encouraging and supportive to the point of making sure you do not ignore directives one and two. So I'm getting long here, so let me move on. I think I've made that point as clear as I can. Lastly, semantic AI search. And yes, behind the scenes video, I'm again using the air quotes. That's the terminology in the industry. What does that mean? A hopeful return to normalcy of language use in social media and across the entire internet. When social, I've got to give backstory, sorry, but details matter. So the backstory, I go back to dial up modems. Before the days of the internet interconnectivity, my computer would dial up another computer and we would talk and do things and so on and so forth with other computers. So I go back before the internet was even born, before I've been on social media since it was invented. I'm old, I'm 63. So, but when social media or till social media arrived, Google used as it became the primary number one search function. I I use I love it for shopping. It's great for finding great rates on widgets, but I don't use it for serious information research searches because it has bias in it, allegedly, as if you don't all see it. Like my articles, uh you could type the name of one of my articles directly quoted into Google, and guess what? It's not likely to show it to you. So don't talk to me about alleged bias. So it filters, it sorts, it makes determinations what it's going to elevate, and money talks. Certain things buy their way to the top of the list, and others, based on its bias in its coding, also filter up to the top. Other things filter to page number one million that you will never get to. Now, maybe my article is there on page one million. Whoever really goes past page 10, you probably don't go past page one or two in your search results, regardless of the search engine you're talking about here. But Google and the other search engines implemented keyword usage, the forerunner to the hashtag. We didn't define the keywords, the systems looked for patterns and trends. And the word, let me use one of my hashtags that I coined, melee at the LCA. For those of you who are basketball fans, you may remember a few years ago, Stuart of the Pistons got into a dust up with whom I coined and hashtag LaMoron James. Well, years before that, at the palace, there was a dust-up, a full-blown fisticuffs that broke out at the palace of the Auburn Hills. Somebody flagged and coined malice at the palace. Well, Stuart and LaMoron James dust up, I coined melee at the LCA. The same kind of flow as malice with the at the palace also begins with an M. Melee, same in a way as the malice at the palace thought. Anyway, so melee at the LCA. If the search engine starts to see that term used all over, that would have become, but you know, I'm talking before this happened, as a keyword. And it would then start looking for that keyword and related items based on that keyword. If I writing an article regarding Melee at the LCA, would then also relate to other pieces from other people, talk shows, radio shows, whatnot. Because yes, I've been on Stone Cold Sports Truth on whamradio.com, live button, link there, Derek Sports Shows on Sundays discussing it as well as other shows. But keywords, I don't want to drag this out. Keywords then became hashtags in social media. Somebody thought, why should we let the system decide the keyword? Let the people decide what is the keyword or words in their posts. Hence the hashtag was born. Hashtag melee at the LCA. Well, now the trend with what's called semantic AI search. AI is smart and quicker. We're going back to the system, recognizing patterns and keywords that keep popping up in posts over and over. You may have noticed this on X. Under the trending column, less unless you see an actual hashtag. You will see a word in plain English or a short phrase in plain English shown as a trend. A lot of people are mentioning this word or this phrase. It is a trend. It is a truer trend. The system is now smart enough to recognize that without people flagging an actual hashtag to get it noticed. So now old dog gotta learn new tricks. I've got to stop using so many hashtags. Had that conversation on X with somebody who ended up blocking me because they don't didn't like all the hashtags I kept putting and things. But that's the way the internet went to. But now we're going back to the system smart enough to recognize the trends of language. So we can write our posts in regular language without so many hashtags in that blazing blue, disrupting our reading flow. So that's in a nutshell what semantic AI search is and what it is doing as the latest evolution of social media and the internet. So we can, could, and should, again, old dog got to learn a new trick. I've been in social media since it first began, so I'm used to using a lot of hashtags to get it cross-referenced. I've got to stop doing that. Because I get it. It is annoying to see a post with a bunch of hashtags, but they were in there for a point and a purpose. Don't need them anymore. Thankfully. So thank you to AI for that. And rock, the XAI team. Big part of leading the way in this trend, it'll move into other systems too. So dial back on the hashtags, do more just regular typing a sentence, like you would if you were writing a letter to someone. Yes, character limitations still apply. Even though X has expanded what a tweet character limitation content used to be, it's still relatively short compared to other platforms. And yes, some I call it the Twitter attention span or the TikTok attention span. They just want. Oh, I can't you tell me in two sentences? Sometimes no. No. A headline can be deceiving. That that article title, that link can be deceiving. Details matter. You have to read the whole newspaper story. A lot of facts are sometimes due to an media. One of my hashtags, hashtag E-N-E-M-E, uppercase, lowercase, D I A, Namedia, biased media, misleading headline. Bury the facts that actually almost contradict the headline at the end of the story. They hope you'll never get to. So online articles, the whole article matters. Details matter. Being fully informed matters. You have to read past the headline. But we can do shorter sentences and longer posts. Unfortunately, a lot of people just don't read. Delusions. They don't know. They don't want to know. They don't care to know. They only want to cling to their delusional bubble. Facts be damned, details be damned. But at any rate, yes, please return to more proper English. Oh, I where I was going with that is even with longer character allowances, more words allowed, the acronym phrase and Grok being encouraging and supporting laughed along with me and agrees. Uh, whether it believes it or not, being encouraging and supporting agrees with me. Initials and acronyms aren't going away. LOL, IRL, BRB, BBL, AWA, AWK, away from keyboard, be back later. Those things aren't going away because people have gotten used to them. And frankly, the human race is getting more and more lazy. So whether they have more space, they're not going to type out B back later. They're going to keep putting BBL. Rather than laughing out loud, they're going to put me to L O L. So the acronyms and the shortcut abbreviations likely aren't going away anytime soon. That's both a good and a bad, and sometimes ugly. To bring full circle back to where all this AI began. There's good, there's bad. There will be ugly along the way. So, anyway, I'm gonna wrap this up. I I don't remember what time I hit record, so I don't even know how long I've gone. It is longer than I wanted to. Thank you for all who indeed understand details matter and actually stay tuned in for the entire broadcast. Thank you all. Take care. Love you all. God bless.

SPEAKER_00 43:28

Same three filters, same four lights. They say this loot is poison, but I don't care.

SPEAKER_01 43:46

I'm doing don't fight.

SPEAKER_03 45:51

Thank you for having tuned in to another Christitutionalist podcast show. I really appreciate that you stop by. Again, please like, share, subscribe. We need you to help spread the Christitutionalist movement. Thank you again. Take care. God bless. Love you all.