Tales & Trails of a Traveling Wiseguy

Earlier today I wrote about how I handled leveling up recently in my D&D game. To make training part of the story without bogging down in minutiae— or, as one player put it, using up group time on four solo adventures— I tried giving the players a storytelling prompt. Think of 1 bad thing (challenge, adversity, setback) that happens and 2 good things. With those elements I'd help craft a narrative for each character, à la the "training montage" device you see in a lot of movies.

I'll share two of the montages we came up with. One came from a great prompt response by the player. The other started with a player who wasn't sure what do to, but when I proposed a challenge, rolled with it and created some entertaining roleplaying moments.

Oh, but first here's an update to the picture of the group leveling up:



In the previous picture, the party's rogue was actually the spy they captured and killed. "Why's the spy leveling up with us?" players demanded to know. 🤣 AI slop. And while I was prompting AI to correct that I figured, "Herran never goes anywhere without her horse...."

Now for two of the training montages.

The Cleric's Unenviable Assignment 

Hawk led off because she had a pretty good prompt response for her adversity. "Kiarana checks in at the temple of her faith... and the high priestess there is a Berenar!"

House Berenar is the ruling family in the realm. The party has gotten itself associated with House Tashara, who are ruling coalition partners. While these noble houses are partners there's also rivalry. Think of it like the Montagues and the Capulets... or, as one player calls it, the Harkonnens and Atreides. In the most recent adventure the party killed a member of House Berenar who organized an attack against them. Berenar couldn't punish them directly for that, since the party's actions were lawful, if grim; but members of House Berenar have learned about it and are hostile.



So Kiarana faced a hostile local leader in her faith. She was assigned scut duties for her training in the temple. That meant, for example, ministering to the poor in Fisherman's Bottom, a seedy district of town authorities are happy to ignore. But that was also really cool because it gave me a place to set the hook for the next adventure, which takes place in Fisherman's Bottom.

The Mage's Strict Tutor

J., playing the mage, was unsure what to use as an adversity element. "Maybe it's hard finding a tutor because they don't like where I'm from," he wondered.

"How about this," I shot back. "You find a tutor easily, and she respects your educational pedigree. She even knows of your first mentor by reputation and is impressed. She adds historical context that you're eager to learn. But she's strict and won't teach you the spell you most want to learn, Fireball."



J. accepted the idea but kept trying to write Fireball on his spell list. I kept reminding him that his tutor refused to teach him that spell. I had fun roleplaying this impromptu NPC as something of a female Severus Snape. She snapped at J. not only about the fact that he'd learn Tiny Hut instead of Fireball but verbally slapped him each time he acted indecisive or forgetful.

BTW this picture I created on my own, no AI, using a found image for a stern-looking woman wizard and adding my own text.

BTW2, when I was roleplaying with J. his harsh task-mistress I didn't just snap at him in character. At one point when I proposed a self-serving trade to him and he worked out that it was an unfair deal, I coached him that he could push back against her. "No," J. said dejectedly, "That'll just make things worse." 😂

Sunday's D&D game wasn't finished after the epic chase scene where Leoghnie and Otonio caught the fleeing spy/ringleader. For one, there was also action happening in the other ring. Back at the scene of the ambush, the hobgoblins had fled but the worgs (bigger, stronger, evil wolves) had not. The worgs had already killed two of the caravan's horses and were about to kill two more. Ryuu-Han the mage moved around from behind his wagon to confront the worgs.



Ryuu-Han's move was a good one... initially. He gained clear line-of-effect on the worgs with his spells while remaining 20' away from the melee. He struck one worg with a burst of magical flame that singed its hide. But in the heat of battle he forgot that worgs, like their wolf cousins, move fast. The worgs disengaged from the horses and moved to attack the foe they perceived as the biggest threat. Ryuu-Han was flanked.

The mage wasn't done. He still had fight in him. The magical protections he'd cast gave him some defense against the worgs' snapping jaws— much better than the worgs expected for a city-ape clad only in rags. His strongest spells were depleted, though. The risk was very real that the worgs, even both injured badly already, would kill him before he killed them.

His companions rallied to his side.

Herran, already on the ground with her bow in her hands, spotted the trouble first. Employing a tactic of "walking fire"— move, shoot, move, shoot— she closed on the worgs, each shot becoming more deadly as she shortened the distance. Before her first move, though, she banged on the side of a wagon asking Wataru for help.



Wataru, a grizzled older driver, had put up a stout fight so far. As driver of the first wagon he'd already taken a lot of arrows, though. He'd retreated into the cover the wagon to drink a healing potion. Now it was time to rejoin the fight. He leapt down from the seat, bow in hand, and followed Herran with walking fire.

BTW, if you're wondering why I'm narrating that Wataru had a bow in hand, when the picture above clearly shows him ready to draw his sword: it's because this is the sort of thing I get from AI image generation. Yes, I'm providing a lot of careful prompts. No, it's not perfect; it even stubbornly refuses to change some details I'm explicit about. I consider it enough of a victory that it doesn't draw any of the female characters with exposed cleavage or bare midriffs (IYKYK).

Update: I worked on a picture of Herran rallying aid to help Ryuu-Han:



Another driver joined in the fight, too. It took some urging from Kiarana the cleric, who was seated next to him on the buckboard. Blushing with chagrin that a cleric had to tell him to get back in the fight, he loaded his crossbow, fired, and loaded and fired again.

Together, between Ryuu-Han's resilience under threat and three warriors working their bow, the group defeated both worgs.

But even there, the story of the ambush wasn't complete. There was still the captured spy to deal with!



Kiarana healed him enough to bring him to consciousness. She cast a spell to compel him to speak only truth. Otonio led the questioning while Leoghnie scowled holding her massive sword, providing the implicit "Or else".

Again re: AI image gen... Yeah, it's not accurate that the cleric has a sword. And she's over-armored. But again, that's way better than her showing off her boobs and midriff. Plus, Jill, who's athletic herself, appreciates that Leoghnie is drawn not only with appropriate gear to go fight snarling monsters toe-to-toe but is drawn stocky enough to actually wield that huge sword she's got (IYKYK).

The spy gave up almost nothing in the questioning. He couldn't lie, due to the spell, but the spell couldn't force him to elaborate. He spat and called them names.

Otonio challenged the spy that he looked familiar. "I have seen you in the city. You serve House Berenar."

"Yeah, Tashara kid, you probably have," the spy sneered. "Best you let me go or you'll join me in hell when they find out."

Otonio's eyes narrowed as he weighed the political dimensions of this. Skirmishes between ruling houses were not unheard of but had to be navigated gently. Very gently.

"Get the merchant," Kiarana barked. She was irritated by the spy's stonewalling. Plus, now it seemed the spy might actually be the ringleader. Leoghnie went and grabbed the tied-up former caravan leader, Munetoshi, dragging him ungently across the ground to add to the questioning.

"Look, here's the deal," Kiarana explained to the two captives. "One of you talks, one of you loses his tongue. Choose." Eyes widened around the gaming table that the cleric was the one making the threat. "Hey, I'm Lawful Neutral!" the player rebutted.

Munetoshi babbled that although someone else had hired him first, this man, Touichi Berenar, told him there would be an ambush and paid him extra to advocate for surrender when it came. This confirmed that Touichi was not merely a "spy" but the coordinator of the attack.

"Hmm," Otonio said aloud, making a show of thinking about the situation as he faced Touichi. "You planned and attempted my murder. I am sure that my loyal house guard would object to that, strongly."

Leoghnie was busy picking her nose, or something— but picking it fiercely— so she didn't catch the verbal handoff. Otonio repeated himself, "I said, I am SURE my house guard would object to that, strongly!" Leoghnie didn't need a third invitation; she lopped off Touichi's head.

"As for you—" Otonio said, turning to caravan leader Munetoshi and leveling his rapier at the man's throat, "Your treachery imperiled all of our lives. It killed one of the men you hired. And should we further risk our own lives spending our now-depleted resources transporting you to the next city for trial? As a member of House Tashara, ruling coalition of Durendal, I declare your life forfeit here and now."

Last night I wrote about finishing off two bottles of booze in my out-of-control collection. I suggested that the next bottle I'd knock out would be San Matias Extra Añejo. Well, that was an oversight. While I really dig that tequila and wouldn't mind finishing it off, the discipline I'm practicing right now is to finish off bottles that are almost empty. San Matias is still over half full. What is/was almost empty? Espolón Añejo.



I poured the last shot and a half for a nightcap drink on Friday.

With this I have now cleared out three bottles this week. Three! Three bottles of booze. It makes me feel like this:



OMG, I can't believe Google Gemini let me make a Sesame Street cartoon about booze! 😳🤯🤣

I am sure that 6 months ago it would have refused, offering me an explanation about copyrighted property, blah-blah-blah. Our future AI overlords are getting bolder.

Across the last two sessions of my D&D game the group sniffed out treachery in their midst. They'd embarked on an adventure as guards on a caravan transporting high-value goods. On their second night on the road, their first night camping in the wilderness between towns, they were attacked by a pair of worgs.

Worgs are like wolves that are slightly tougher, much smarter (they can learn languages!), and evil.

Fortunately for the caravan, Leoghnie, the toughest warrior in the group, was on watch at the time. She killed one and chased the other off while waking her compatriots.

In the morning the four PCs and their NPC ally, Otonio, compared notes. Otonio had already warned them that "Every thief in Durendal knows about this caravan, so we should expect trouble." Ryuu-Han shared that in chatting with his watch-mate last night he learned that most of the guards were noobs. The crew chief was offering shockingly low pay for this being a conspicuously high-value caravan. Otonio amplified that, explaining that he was asked to provide a few "regular" guards from his household before he surreptitiously hired the PCs instead.

Kiarana decided it was time for some divine assistance. She memorized a few Zone of Truth spells at dawn and called the crew together for a team meeting after breakfast.



Everyone, including the PCs, was quizzed with the same questions about knowledge of the caravan being attacked. All of the hirelings disavowed knowledge or involvement in any treacherous plans. Only the lead merchant, Munetoshi, refused to answer the questions directly. He gave evasive answers and complained repeatedly about the disrespect inherent in the process. "Mutiny," he called it, twice. He tried to leave but well-armed and armored Leoghnie blocked his path.

Although Kiarana cast the spell and started the interrogation, Otonio took over as he was more fleet with words. "Your evasiveness isn't a good look," he explained to his boss, Munetoshi. "Your inability to answer simple, direct questions with a 'No' under the cleric's Zone of Truth spell leads us to the conclusion you made treasonous plans against this mission."



The group decided to take Munetoshi prisoner. Otonio got some ropes from the supply wagon and tied him up. Then Leoghnie's player realized, "Hey, I have Rope Use +6" and said she'd tie him up extra good.

About these pictures: I created the latter two illustrations with Google Gemini. I prompted the AI with a variety of information about the group and the situation. I iterated a few times to try to improve some of the details. Some of those prompts worked, some were ignored or only partly heeded. And some things are just funny, like the fact that each picture contains a person with three arms— and which person it is changes from pic to pic!

Keep reading: The real ambush comes hours later!

I've been writing about a D&D adventure I created and DMed recently, The Collector's Menagerie. I shared in my last blog that a player noted the names/types of rooms in the mansion setting— The Hall, The Library, The Conservatory— and remarked, "This is like a game of Clue!" And how I quipped, "With monsters lurking in the rooms waiting to kill you, it's like Cursed Clue!"



You might wonder, given the setting in a dead guy's mansion and the (twisted) murder mystery element to the story if I conceived this adventure as, "It's like Clue, but things in every room are trying to kill you." Frankly it would be awesome if that's how I came up with it. Alas, I did not. Not this time.

I have, in the past, created memorable adventures that started with the simple idea, "What if X, but also Y?" Or to be more specific, "What if something culturally familiar to us players in the modern day were the setting of a swords-and-sorcery fantasy story?" and fill it with in-jokes to see how soon the players figure it out. My greatest hits in that vein have been "The heroes traverse a magical Gate to a Renaissance Faire circa 1995 (pre-cell phones) but think it's actual early Renaissance" and "All the traps in the lich's lair form the lyrics to The Eagles' Hotel California." 😆🤣🤘

Yeah, it could have been epic if I started with "Cursed Clue". But I think it is kind of epic even though I only kind of backed in to the story being Cursed Clue.

My kernel of an idea for this adventure was simply, "Monsters are in a city mansion". I used AI to flesh out the idea. That got me to the point of it being a variety of exotic monsters (read: magical beasts and aberrations) that had escaped their cages after the owner of the house, a reclusive collector, died recently.

For the mansion itself I already had a map of an actual English city mansion I'd used as a setting in a previous game. I grabbed that to use again here. The names of the rooms on the map reminded me of a mansion map I know well from my childhood....



Yes, Clue! And it was because of the maps that I made the connection. The real-life mansion floor plan had rooms marked Hall, Ballroom, Conservatory, Drawing Room, etc. Those reminded me of the rooms in Clue. BTW, the Drawing Room is the Lounge. The terms are basically interchangeable in historic wealthy Western homes, indicating room a full of lavish but comfortable furniture for withdrawing to after a meal to impress guests.

Once I made the connection myself I thought about how to lean into the idea of "This is Cursed Clue". I tried to think of a way to stash treasure items, some analogue of the candlestick, rope, knife, etc., in various rooms that the heroes would need to recover to complete the challenge. Ultimately I punted that because it seemed too complex. Simplicity was one of the things I was after with this adventure idea. But I did put in some ridiculous secret doors connecting rooms on opposite sides of the map. Shh, the players haven't found those yet!

2025 was the Year of AI. Content generated by AI started popping up everywhere. I even used it a bit myself. But just a bit, because one thing that was blindingly obvious in the Year of AI— obvious to anyone really paying attention, anyway— is that AI can produce some laughably silly results.

For example, some of my colleagues went big into using AI image generation to illustrate slide presentations a few months ago. I couldn't help but point out, publicly, when a person (in the picture) who was the subject of the story had, say, 3 arms, or when they had a laptop showing our product on its screen while the screen was bent around the wrong way. Maybe that makes me a bad person. But I've always been the one dumbass who, when the emperor strides onto the stage stark-fucking-naked, nudges people next to me and says out loud, "Look, the emperor's wearing no clothes!" And more to the point here, if we don't object to AI slop right now, its' going to become normalized and we're going to be completely inundated with it in 2026.

Anyway, this journal entry isn't supposed to be a screed against AI. I'm writing to share some in-the-know humor about some of the funny results AI image gen gives us.

A few months ago I used Google's Gemini image gen to illustrate panels for a story I wrote on my blog. It's the one I only finally finished yesterday: The Mystery of the Church Up the Hill. One of the images I created was of my father painting the inside of the church he attended decades ago. I wrote a prompt like



And the first result was....



Disco Jesus! 🤣

I literally gave this prompt next:



To its credit, the image generator came back with a new image that did not have the son of god painting his own church after rising from the dead inside a vintage clothing shop. 🤣

Ultimately there were more things wrong with the pic than just "My dad doesn't look like Jesus", so I prompted the AI to start over. On my second try I used a few more terms to describe the aspects of the scene I thought were most salient. I got the image I used in the story I shared yesterday.



Was the final image I went with at all like the church my dad painted? No. But it conveyed the parts of the story I thought were appropriate. Including a few key elements of my dad's appearance: age, body shape, hair color, and glasses. One thing I couldn't get right in a handful of prompts was managing to dress my dad like a dork from the early 70s. Gemini kept taking the "early 1970s" prompt as making my dad look like a dork who dressed up to go disco dancing. Though I can see now that Dad would've looked pretty sharp— for a dork— in Chelsea boots and a leather vest!

It was only a matter of time. For a few years we had the scourge of robo-spam calls. You know, spam campaigns where a recorded voice tries to trick you into something, like your relatives in China are in trouble with the central government, and asks you to stay on the line for more instructions. (Yeah, that one was easy to detect— and ignore— because it was in Chinese and I have no Chinese relatives.) But now, because it's 2025 and AI is popping up everywhere, we have AI robo-spam. Spammer aim to increase their hit rate— the chance that you interact with their pitch— by making it seem almost like a real person is talking to you.

I answered my first call like this last night.

Note, I don't answer many spam calls anymore. They've become easy to spot, as the Caller ID comes up "Potential Spam" on my smartphone. I chose to answer this one on the theory of once you answer they stop trying. (I base this on knowledge of how auto-dialers work, from my experience as a telemarketer 😨 many years ago. The system will keep trying your number periodically until it logs a live-person connection.) Plus, sometimes these calls are not from spammers, per se, but from organizations I have a relationship with.

The call began innocuously enough, with seemingly a live person on the other end.

"HI, IS THIS <YOUR NAME>?"

"Yeah, this is <first name>."

"<long pause> THIS IS ERICA FROM CREDIT-SOMETHING. HOW ARE YOU?"

The voice seemed a bit off. It was natural sounding but boomy. And it was too perfect. That triggered my suspicions. Most spam callers nowadays frankly struggle with English— because they're low-skill workers in foreign countries where the cost of labor is lower than in countries where English is the primary language.

"What's this about?" I asked, aiming to short-circuit the obvious cold-call.

"<long pause> GREAT! I'M CALLING ABOUT—"

The second long pause and the fact that the caller responded as if I'd answered her question ("How are you?") instead of countering with my own question told me I was likely speaking to a robot. And by "robot" I mean an AI powered system. Though obviously not a great one.

"Are you a robot?" I asked, interrupting

"<long pause> I AM AN INTERACT VOICE ASSISTANT!"

"Ergo, you're a robot. <click>"

At my sales training seminar the past few days I had a number of conversations with colleagues about AI. These convos spanned topics from "What are we doing [in our product] to align with industry demand for AI powered features?" to "How can we use AI in our jobs in sales to sell more effectively?"to "Is our job [in sales] even going to exist in 5 years?" There's so much I could write about AI even within these topics, let alone the broader discussions about AI. For this, my first journal entry about AI, I'll start with the latter question— which, to state it in more dire terms, is,  Is AI coming for my job?

I use this alarmist language to make a point: This is what people are worry about more and more. And this is the type of lanaguage that's becoming increasingly common as people express their thoughts/concerns.

I don't think the future is as bad as all that. I think we're at a point in the technology hype curve where there's a lot of uncertainty. And I want to be careful to say that I really can't predict the future of AI, even 5 years out.

Why 5 years? Consider how far AI has come in 5 years. 5 years ago AI was more science fiction than science fact.

Three years ago AI was full of hype but still short of reality. While many people in software development, my field, were buzzing about how AI would give us "10x" improvements and pouring money into it, a few of us were pointing out that there was currently no there there and such investment was like the proverbial lemmings chasing each other over the cliff.

Two years ago in software development we started to see the actual value of AI appear. AI could write code— but generally simple code, and it needed more testing and definitely review by a skilled person. The new wisdom became, "AI makes programmers 30% more efficient." That's a far cry from the 1000% gain people were still frothing about 12 months earlier!

Today, in software, we're seeing that 30% level of gain take hold more broadly. Some people react to that figure by asking "Does that mean layoffs of 30% among software developers?" I think that viewpoint fails to appreciate what's happened across the history of technological progress.

Yes, new technology has always reduced the number of old jobs that were doing things the old way. In the industrial revolution factory automation reduced the number of jobs for everything from sawing wood to stitching clothes to digging for coal. A simplistic view of it is, "Machines replaced people." But while machines replaced jobs where people were doing rote, manual work, the economy was not a zero-sum game. Overall the economy grew because of efficiency, and new, higher value jobs were created elsewhere.

The same lesson applies with the AI transformation. AI will replace people who are doing lower level, more rote jobs. But economic gains will mean more higher level jobs can be created elsewhere. For those who are looking at it as zero-sum, though, and wondering, "Will AI take my job?" the answer is really, "People who know how to use AI to be more productive will replace those who don't."