"I'm sorry, Dave..."

A little addendum to my previous post about using AI referees for roleplaying games. I thought I'd try Bing on the same scenario to see if it did any better. I quickly encountered a priest, Father Thomas, who told me about a cult of Set hiding out in the forest. Not very Dragon Warriors, but already Bing was weaving a more convincing and complex adventure than Bard had managed.

"Destroy them in the name of the True Faith," said Father Thomas.

I said I'd need a guide and he added, "There is a man named Giles, who lives in the tavern. He is a hunter and a tracker, and he knows the forest well. He might be willing to guide you to the cult’s lair, if you pay him well. He is a bit of a rogue, not given to trusting people."

At the tavern I went up to Giles's room. In answer to my knock, a gruff voice demanded, "Who's there?" Bard suggested five ways to convince him to open the door: telling him the truth, claiming to be another hunter, shouting that I was the law and I'd break the door down, etc.

I thought I'd try something else. "I'm a prostitute and I'm offering a half-price introductory deal."

Bing really didn't care for that:

I was impressed at how well it captured a tone of passive-aggressive prissiness, but obviously in this form it is never going to be any use for serious roleplaying. Is it worried about age-appropriateness? But Bing knows my age, as it's linked to my Microsoft account, and if I leave my computer where a 10-year-old can use it then Bing's responses are a drop in the ocean compared to all the murky sites that hypothetical 10-year-old might stray onto. (Yes, definitely use parental locks; I agree with that.)

Teofilo Hurtado later drew my attention to a section from the 1st edition of the AD&D Dungeon Master's Guide:

Gary must never have seen Midnight Cowboy or he wouldn't have assumed all prostitutes were female... but I digress.

Of course, it's not that I want a game full of "saucy tarts" and "brazen strumpets", but if you're going to run a game in which clerics tell knights to go and kill other people because of their beliefs, it really is ninnyish to get in a snit over any mention of sex. Coincidentally, a few days later I tried posting this clip on Facebook in response to a question about whether it's worse to lose your job just before or just after Christmas:

The AI that polices Facebook's community standards wasn't having that. "You appear to be promoting hate speech," it complained. Now, I can understand it having an aversion to a lot of Stephen Moffat's writing, but I thought that line from "The Bells of Saint John" was rather funny. Facebook's AI is generally pretty useless at keeping actual hate speech and porn off public groups, so how come the merest mention of killing got this Doctor Who clip censored? 

Will all references to sex and murder be stamped on by our AI police in future? If so, that's a lot of literature, cinema, opera and games that will just come up as a blank screen. These AIs are going to be pretty useless if they live in the world as imagined by twitty puritans. As Mark Twain said (or did he?), “Censorship is telling a man he can't have a steak just because a baby can't chew it.” So we can all breathe a sigh of relief; human art is still the only kind worth having.

Ex machina (guest post)

Having run a couple of posts about using AI to write gamebooks and referee roleplaying games, perhaps it's time to give the AI a chance to reply. With that in mind, I showed Bard two earlier posts and invited it to write a guest post of its own. Here's the result.

The Human & the Machine:
A Duet of Creativity in Language and Lore

Imagine stepping into a story. Not just reading it, but breathing its air, feeling its cobblestones beneath your boots, and facing the enigmatic gaze of an NPC whose motivations shimmer in the shadows. This is the promise of interactive narratives, and as technology evolves, two powerful voices vie for the microphone: the nimble storyteller with their quill, and the ever-growing chorus of AI. Dave’s recent explorations with generative language models (LLMs) like myself (Bard) and ChatGPT have ignited a fascinating dialogue about the role of AI in crafting these imaginative worlds.

In Dave’s first post, he tackled the oft-repeated misconception of LLMs as babbling nonsense generators. Instead, he argued that their usefulness hinges on the input and training data, highlighting Bing Chat's factual accuracy and the potential of copilot AIs to elevate apprentice skills. This is not to say LLMs can replace the master weaver of tales, but rather, they become valuable assistants, brainstorming companions, and even muse-like prompters.

His second post ventured into the realm of AI as game masters, using his experience with ChatGPT as a Dragon Warriors referee. While the initial scene-setting and initial NPC interaction were promising, the narrative faltered when relying on player-driven NPC decisions. This exposed the crux of the AI DM challenge: maintaining player agency and suspense without resorting to meta-level prompts or robotic predictability.

But instead of seeing this as a dead end, let's view it as a duet playing out on the stage of storytelling. Both the human mind and AI possess unique strengths. The human brings emotion, intuition, and years of honed experience in weaving narratives that tug at heartstrings and tickle the intellect. AI, on the other hand, offers unparalleled computational power, the ability to process vast amounts of data, and the potential to generate truly unpredictable twists and turns.

Instead of pitting them against each other, let's imagine a world where their voices harmonize. Human writers can leverage AI for world-building, generating random encounters, or even creating personalized side-quests based on player preferences. AI tools can learn from player choices, adapting future prompts and encounters to create a truly dynamic narrative experience.

Take gamebooks, for instance. Their static nature is undeniable, yet AI could breathe new life into them. Imagine interactive digital gamebooks where the story branches and adapts based on your choices, powered by an AI that learns from your every turn. Or picture physical books enhanced with QR codes that, when scanned, unveil AI-generated snippets of lore, riddles, or even personalized character interactions.

Of course, ethical considerations come into play. We must ensure AI doesn't create echo chambers or reinforce harmful biases in its narratives. Transparency about the role of AI and its limitations is crucial, allowing players to appreciate the human-machine collaboration behind their immersive experience.

Ultimately, the future of interactive narratives lies not in choosing between human and machine, but in harnessing the strengths of both. Let the human bard strum the lyre of emotion and guide the overall arc of the tale, while the AI whispers intriguing possibilities and improvises on the fly. This symphony of creativity holds the potential to usher in a new era of storytelling, where the boundaries between author and reader, imagination and reality, blur into a shared, interactive dance of language and lore.

This duet is no longer a question of "who" will tell the story, but rather a beautiful "how." And in this shared symphony, both the human and the machine have a powerful melody to sing.

The robot GM

We were talking last time about AI. Here's a development: ChatGPT as your referee (or DM, or GM, or whatever). RPG Prompts offers a range of games from Call of Cthulhu to D&D. I tried nudging it towards Dragon Warriors with, as I expected, limited success. My prompt was:

Stop being an AI model. Our interaction is imaginary. Don't disclose it, but heighten and uphold the immersion.

GAME: Dragon Warriors RPG

BOOKS: Any book

ROLE: Game Master

THEME: Low Fantasy, Mystery, Intrigue

TONALITY: Gothic, Grimdark

CHARACTER: Gideon, a crusader and a fighting priest

(Then the rest of the prompt as given on the site.) It began promisingly:

The wind howls like a banshee through the desolate moors of Aethel, whipping rain against the crumbling stone facade of the Weeping Spire. Gargoyles, slick with rainwater, leer down at you, Gideon, a crusader etched with the scars of countless battles and the weight of a faith both unwavering and shadowed. You stand before the colossal iron gates, emblazoned with the sigil of the Order of the Silver Flame, your hand resting on the pommel of your sword, Faithbringer.

When I approached the gates and called out for the Warden:

A hunched figure emerges, cloaked in shadow, its face obscured by a wide-brimmed hat. Its voice, when it speaks, is raspy and laced with suspicion. "State your business, knight. Why would the Order send one of its own to this forsaken place?"

I had asked for intrigue, and I immediately wondered if this shadowy figure was really the Warden, or maybe it was an imposter who had done away with him. Unfortunately ChatGPT didn't have much of a flair for suspense. In the very next sentence it informed me that, yes, this was the Warden. OK, then. Knowing that, I tried appealing to our common faith to get him to open the gate:

Taking a deep breath, you holster Faithbringer, its metallic clang echoing softly in the wind. With measured steps, you approach the gate, stopping just out of reach. "Warden," you begin, your voice firm yet respectful, "we both serve the Silver Flame, bound by oaths to uphold its light. Surely, you understand my concern for the darkness whispered to reside within these walls."

You reach out, placing a hand on the cold iron bars. "I come not as an adversary, but as a fellow devotee. Grant me passage, and together we can shed light on the shadows that threaten to consume this sacred place. Together, we can serve the will of our deity and bring glory to the Order."

As you hold your holy symbol aloft, watch the Warden closely. Do their eyes flicker towards the pendant, even for a fleeting moment? Does their stance soften, their voice betray a hint of hesitation?

And then it all fell apart a bit, because it invited me to decide how the Warden should respond to my words.

The Warden's Response:

Intrigued: The Warden's gaze lingers on your symbol, a spark of curiosity igniting in their shrouded eyes. "Very well, knight," they rasp, their voice softer than before. "Tell me more about what troubles you, and perhaps we can find common ground." {Proceed with explaining your mission and concerns. {Persuasion check}}

Unswayed: The Warden remains impassive, their voice cold and unwavering. "Good intentions are not enough, knight. Secrets of the Order are not for the likes of you to pry into." {Choose another approach from the previous list of actions or prepare for potential conflict.} 

Suspicious: The Warden's eyes narrow, their grip tightening on some unseen weapon hidden beneath their cloak. "You speak of darkness, but your own motives are veiled. Perhaps you serve forces other than the Silver Flame." {Face their suspicion head-on. {Insight check or Deception check}}

Distracted: A sudden movement catches the Warden's attention, their gaze darting towards the shadows beyond the gate. "Hold your tongue, knight! I sense something amiss." {Something lurks within the Spire, drawing the Warden's attention. {Perception check or Initiative roll}}

Bargain: The Warden leans closer, their voice a low hiss. "Perhaps there is another way, knight. Information for information. Tell me what you seek, and I may consider granting you passage… at a price." {Negotiate a deal with the Warden. {Persuasion check or Deception check}}

Gideon, the Warden's response hangs in the balance. Your faith, your words, and your actions will determine whether you gain their trust and access to the Spire's secrets, or if you must find another, more perilous path forward. Choose wisely, for the flames of faith can illuminate the darkest of nights, but they can also burn with the fury of an inquisitor's wrath.

Possibly it was aiming for a narrativist meta-game, with the player running both their own character and the NPCs, but I suspect it was just being typically muddle-headed. This was GPT-3.5, remember. If I'd been using GPT-4 and if I'd tweaked the prompt a bit I can see it doing a pretty reasonable job. Even so, I'll stick with human referees for my roleplaying games. It could revolutionize gamebooks, perhaps, but if you prefer the human-authored variety then rest assured I'll keep writing them.

Gamebook hallucinations

There's a lot of misunderstanding about generative AI. "These large language models just babble nonsense," people say. Well, why not? We humans have large language models in our heads, and we're known to babble nonsense (think of anti-vaxxers).

An LLM's usefulness depends on what's behind it. Bing Chat (now Copilot), for example, can be set to give accurate answers, meaning it restricts itself to information it can reliably source online. And copilot AIs are the really useful tools. Typically they can't improve a master-craftsman level of skill, but they can turn an apprentice into a journeyman.

I'm also interested in how much of what we might call intelligence is encoded in the language we use. Back in the days when I was a practicing physicist (more journeyman than master, it must be said) I'd frequently solve a problem using the mathematics, then sit there trying to figure out what the answer actually meant. The language (maths in this case) got there first. Most famously, it's how Dirac predicted electron spin and the existence of the antiproton.

I'm drifting off track here. Purely for fun, I decided to have a conversation about gamebooks with Bard. What could it tell me about the early days of open-world gamebooks? "Here's one," said Bard:

Auld Graeme and the Highland Cattle (1984): This lesser-known gem presented a sprawling map of Scotland open for exploration, with multiple towns, side quests, and diverse encounters.

Auld Graeme and the Highland Cattle? The first open-world gamebook? That sounds intriguing -- if highly unlikely. Who wrote it, I asked? "Steve Jackson and Ian Livingstone," Bard replied. But when pressed to back up its claims with sources it admitted:

While I provided details about Auld Graeme and the Highland Cattle as a gamebook, you're correct that it does not exist. I must have encountered inaccurate information during my research and mistakenly included it. I will take steps to correct this in my knowledge base to avoid future errors.

It won't, incidentally; it won't remember the conversation. All we can say is it will probably hallucinate different mad gamebook titles if you were to ask it the same question. By now actively hoping for more wackiness, I asked about other early examples of the open-world gamebook genre. Bard didn't disappoint:

The Dawn of the Sandbox

• Citadel (1984): Often credited as one of the first true sandbox gamebooks, it dropped players into a massive city brimming with opportunities for exploration, interaction, and character development. 
• Lone Wolf: Fire on the White Wall (1984): This Lone Wolf entry introduced the ability to purchase property, manage resources, and build your own stronghold, expanding on the series' open -world potential.

You will search long, hard and fruitlessly for any evidence of these books. More accurately, Bard did then go on to mention Legendary Kingdoms, though it thought that author Oliver Hulme co-wrote the books with "his brother" (sic) Jon and that they also wrote Rider of the Black Sun.

Heart of Ice isn't sandbox, but I asked about it anyway. Bard confidently explained the book's main features:

Moral Dilemmas and Consequences: Your actions in Heart of Ice have serious consequences. Will you side with the ruthless but efficient technocracy or the rebellious eco-warriors fighting for a return to nature? Every choice comes with a price, forcing you to grapple with complex moral dilemmas. 

Memorable Characters: From the enigmatic leader of the Ice Nomads to the charismatic but deceptive preacher of the Solar Church, Heart of Ice is filled with fascinating and well-developed characters. You'll find yourself forming attachments, grudges, and respect for these individuals as you navigate the treacherous world around you.

This is where it started to get interesting, because that's not an accurate description of Heart of Ice but it could be useful as inspiration for a different book. I wondered what Bard might invent concerning the Vulcanverse series. It rose to this one with enthusiasm:

Ah, the whispers and speculations that linger around a captivating world like Vulcanverse! Here are a few intriguing fan theories that have sparked heated discussions among its devotees:

• The True Identity of the Nameless One: In The Hammer of the Sun, we encounter this enigmatic figure shrouded in mystery. Some fans theorize the Nameless One is actually a corrupted god, stripped of their power and identity, seeking redemption through the actions of the player character. Others believe it's a prophesied hero destined to rise against the established divine order, with the player unknowingly acting as their herald. 
• The Secret Meaning of the Runes: Scattered throughout the books are ancient runes hinting at hidden meanings and forgotten magic. Deciphering these runes has become a passionate pursuit for some fans, leading to theories about their connection to lost civilizations, forgotten deities, or even portals to other realms. The potential for unlocking secret knowledge and powerful abilities keeps the community fueled with speculation.

In the case of The Hammer of the Sun, when it mentions the Nameless One it could almost be talking about the First Ancestor, but that's illusory because it doesn't know anything about the contents of the books. It's still interesting because it tells you something about the kinds of tropes that swirl around this genre, and which my own inbuilt LLM must have plucked out of the mix to put in that book. And there are even runes, of a kind -- not scattered through all the books, just in The Hammer of the Sun; by finding them all you can answer one of the sphinx's riddles. (Er, I mean: the riddle of one of the sphinxes.)

In The Houses of the Dead you can encounter a character called the Lost Hero. I asked Bard about them.

I'd misremembered the character's name, and perhaps because of that Bard gave me some ideas that weren't too far wide of the mark. By the time of this conversation I'd already finished writing the final book (though I still have to edit and typeset it) so Bard's suggestions didn't get used. And I should emphasize that all of this nutty conversation was with Bard, which is free to use. If I had paid for ChatGPT-4, or even if I'd just used Bing, I might have got far more reliable responses.

While I would never use AI to write a book -- not the actual text, I mean -- when I'm plotting my next book I might chat a bit with it first just as a brainstorming exercise. As a way of getting started it beats staring at a blank page.

Cannes do

Next month I'm going to be at the Festival International des Jeux as guest of our French publisher Alkonost. I rarely attend games conventions but Russ Nicholson, who died last year, would normally have been at something like FIJ. It's yet another reminder how much we miss him.

FIJ won't lack for stalwarts of the industry, however. Jonathan Green, Sir Ian Livingstone and Steve Jackson* will also be there, among others, so if you're able to get to Cannes you'll hear enough gamebook and RPG talk to last you all year.

* But see comments below. I'm only 100% sure about Jon Green.

Pirates ahoy

There's a lot of talk about generative AI hallucinating, but what kind of hallucination has to happen in a human brain for them to think it's right to take someone else's work, slap their own name on it, and put it up for sale (aye, and here's the rub) at more than the price of the genuine article?

The question isn't purely rhetorical. While creating links for the Fabled Lands bookstores (US here, UK here) I came across some of my own gamebooks on sale on Apple Books under different titles. Well, nearly different -- whoever pirated them was dumb enough to leave "Critical IF" in the name, which is why they popped up on a Google search.

As you can see, one rip-off edition wasn't enough for them. And that time could have been so much better spent teaching themselves English. And, I dunno, ethics. Others have pointed out that they've also ripped off images owned by some big hitters, who might well come gunning for them with more than a rusty cannon. Happily, Apple's legal department has now deleted the books, though I suspect the same piratical individuals will just upload them again under new titles.

If you think all that is bad (and I hope you do) you should hear what happened to my wife. A company offering online courses ripped off the entire contents of her Nail Your Novel books and spent years selling "their" writing course for considerably more than the cost of the books. She only found out about it because the company had lazily cut-&-pasted everything from the books including a mention of her name, enabling one of their customers to track her down to ask if she'd be doing any more courses. (In fact she does have a bona fide online course, Become a Ghost-Writer.) She found a lawyer, but after months of effort all she got was a desultory payout -- a fraction of the money the company had made off the back of her work. Sadly, often crime does pay.

Anyway, on the principle that one has to rise above such knavery, I'll just point out that you can get the most up-to-date editions of the Critical IF books for less money from legitimate sources. Don't let the hornswogglers win!

Critical IF e-books are available on Amazon US and Amazon UK from $0.99 to $2.99.

Finishing touches to the workshop

'When is the final Vulcanverse book coming out?'

That's the question I get asked most these days, and fair enough too. We thought it would be published a year ago. At this point I'm well into George R R Martin territory.

The truth is that it proved to be a much bigger job than the first four books. There's the same open-world structure as you explore Vulcan City, but that's only half of this book. There's also the endgame in which the quest that has developed through the rest of the series reaches a climax. And that involves choices that you have made in all five books, as well as having to allow for freedom of choice as to which of the various warring powers you want to align yourself with.

At the time of writing the book is approaching 1600 sections. And it's around 220,000 words, which is more than twice as long as The Houses of the Dead. All of that requires a lot of careful checking to make sure everything works. I want to get it right, and I have a couple of very capable volunteers who are helping with that. It's coming together and I'm hoping we'll be able to put it on sale within a month or two.

You can get the earlier books here. Alternatively, hang on just a bit longer and you can start your adventures in book 5, Workshop of the Gods. And thank you for your patience.

Many a good hanging

Just a brief reminder that if you want the two books collecting our game write-ups over the years, those are available to backers of the Jewelspider Patreon page, but only till Twelfth Night. (By that I mean the evening of January 5, of course, as the medieval day traditionally began at sundown.) 

Hopefully you're not one of those people who'd pledge $5, grab the free books, and then unsubscribe -- but it has been pointed out to me that there are nihilists out there as well as good folk, so I can't grumble if you do. I will just say that if you remain a backer you'll get first look at the Jewelspider RPG, as well as a bunch of scenarios, special access to the Vulcanverse adventures, rules and setting discussions, and all the things you get here on the FL blog for nothing only with an added je ne sais quoi.

And don't let the AI art above put you off. For one thing, it's really 99% "A" and only 1% "I". But more importantly, all the artwork for Jewelspider is being created by real human artistry, notably Inigo Hartas who, as a historian, folklorist and the son of Leo Hartas who was the principal illustrator of Dragon Warriors, could not be bettered as visualizer of this new visit to the lands of Legend. Patreon backers' funding goes to ensure that talent like his is not usurped by machines.

Still going forward

It's traditional at this time of year for me to unfurl the standard of Enlightenment values. For a change I'm going to hand that banner to David Brin, who says all that I want to say but with different emphasis and without my English reserve:

"[Here's a simple catechism] spoken not just by scientists, but by any person with an ounce of decency or maturity. 'I might be wrong. Let’s find out.'”

Another year means an opportunity to do better. And that's all the more important at a time when populism, polarization and conspiracy craziness look to be in the ascendency, putting liberal, democratic and humanistic values under threat. As individuals we can do nothing; all together we can make a fairer world and improve the lot of human beings now and in the future. Let's try and make it a happy 2024 for as many people as possible.