Game thinking from Adam Clare

Tag: AIPage 1 of 5

Maple Resistance: AI Trump Annexes Canada

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

I built a prototype of a game that explores the takes on of the “bugs” of AI and turns it into a feature. That bug is, of course, hallucinations. The way generative AI works requires hallucinations to function but to the end user the hallucinations can come across as a bug. Hallucinations are what causes the “false truths” and made up facts that generative AI spits out. I started to wonder  if there’s a way to make a game in which the hallucinations are beneficial to the play experience.

I want to create more games about policy and possible futures. In the same vain that the military uses war-games to train we ought to use policy-games to train our politicians and bureaucrats. A game gives us a space to experiment what works and doesn’t without actually causing harm to the real world.

With my AI thinking and interest in policy games I decided to make a game set in the year 2025 in which Donald Trump has won the 2024 election. Since Trump makes up facts and losses track of what’s being discussed he and a simple hallucinating AI could be indistinguishable from one another. The” bug” has become the feature.

The player’s goal is to prevent Trump from annexing Canada, so in Maple Resistance you negotiate with an AI Trump while trying to make political choices to protect Canadian sovereignty.

I also integrated a local LLM in the game; read on for why I made this game and what I learned.

It’s worth noting I started working on Maple Resistance at TOJam back in May and since then the world of both AI and politics has evolved (I was originally going to post this on July 15).

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.


Maple Resistance: navigating annexation through gameplay

In 2025, the unthinkable happens: Donald Trump, re-elected as President of the United States, declares Canada to be part of the USA, following his successful annexation of Puerto Rico. But this time, there’s no military intervention—just a bold proclamation. This audacious scenario sets the stage for “Maple Resistance,” a text and card-based game that explores the intricate dance of diplomacy, identity, and resistance.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Designing dystopia

As a game designer, I like to experiment with new technologies and mechanics. One of my primary goals with Maple Resistance was to experiment with local Large Language Models (LLMs) and Unity to create a narrative experience driven by dynamic, AI-generated dialogue. This worked out. The local LLM worked better than I thought, and given that the LLM I was using is from earlier this year I’m sure that the newer ones are even better. For those that are curious I used Mistral 7B.

AI generated image with the prompt "What Canada would like if the United States of America annexed Canada in the year 2025"

AI generated image with the prompt “What Canada would like if the United States of America annexed Canada in the year 2025”

Card collecting mechanic for conversations

Another experiment was the card-collection mechanic tied to player interactions with NPCs. The idea was simple: engage in conversations, gather information, and earn cards that can be strategically played. While this mechanic had mixed results in its execution, it showed promise. In the “Maple Resistance” prototype, players can collect these cards, though the system is not complete and far short of what I originally envisioned. An interconnected inventory system remains an elusive goal; however, the process has sparked numerous ideas for future iterations.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Futurism and the polycrisis

Set against the backdrop of a polycrisis—a convergence of multiple, interconnected crises—the game explores themes of nationalism, sovereignty, and the fragility of political systems. “Maple Resistance” is a speculative narrative that resonates with current global uncertainties, from geopolitical tensions to technological disruptions. It invites players to ponder the future of nations and the delicate balance of power, all within the framework of an engaging, strategic gameplay experience.

All the cards and dialogue (expect for the AI generated ones) are based on real world instances or statements made in the last few years. The infamous Project 2025 was an influence when I started this back in the spring.

AI generated image with the prompt "A cartoon version of Canadians holding back Americans from entering Canada"

AI generated image with the prompt “A cartoon version of Canadians holding back Americans from entering Canada”

There’s always more

Working on “Maple Resistance” has been a journey of discovery. The integration of LLMs in Unity was a technical success, although the tech I used is already outdated. The card-collection mechanic, while not perfect, offers a compelling layer of strategy and immersion. These experiences have laid the groundwork for future projects that will refine and expand these concepts, ultimately aiming to create richer, more complex game worlds.

I hop that as we navigate an increasingly uncertain future, games like “Maple Resistance” serve as both entertainment and reflection, offering a space to explore the possibilities and challenges that lie ahead. Through games we can create experiences that not only entertain but also provoke thought and inspire dialogue about the world we live in and the futures we can imagine.

Let’s make more policy games!

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

Screenshot of the game Maple Resistance showing dialogue spoken by a fictional Donald Trump with text generated by a local LLM.

 

Play now

Maple Resistance is available to download from Itch.io.

 

 

A bonus image that makes little sense for those of you that read all the way to the end:

AI generated image with the prompt "A cartoon version of America invading Canada in the near future"

AI generated image with the prompt “A cartoon version of America invading Canada in the near future”

Artificial Intelligence in Relation to Games

Artificial Intelligence (AI) has been said by many to bring us a utopia and, now more frequently, a dystopia. Regardless of where research into AI takes us we’ll be seeing the benefits in games in multiple ways. AIs are not new to games and have been used in games for a long time, what’s more is that a good way to test AIs is to use games.

In the 90s an IBM computer beat a world champion chess player and that was impressive at the time. A chess AI can be programmed relatively easy since there’s a set way to play (basically look at all possible moves of a set and pick the best one).

DeepMind

A Game like go is harder to program for and as a result was deemed to be a triumphant challenge for programmers to create a program that can beat a human (the quantity of what needs to be coded for is huge). Last month, Google’s DeepMind beat a top-tier European go player.

Instead of programming for every possible move like in Deep Blue, Google let their program learn on its own. “AlphaGo was not preprogrammed to play Go: rather, it learned using a general-purpose algorithm that allowed it to interpret the game’s patterns, in a similar way to how a DeepMind program learned to play 49 different arcade games.” This is striking because it’s a leap in how we make AIs that play games. We just toss the AI at the game and hope it learns what to do – just like we do with human players.

To hear more about the future of DeepMind watch this lecture by Demis Cassabas (founder of DeepMind) about the future and capabilities of artificial intelligence.

Challenges for DeepMind’s Artificial Intelligence

Does DeepMind seem too good to be true to you? It’s probably because the annoucnemtn around how it beat the go player is a big claim. Gary Marcus deconstructs the advancement and looks at the challenges AlphaGo (and AI in general) needs to still overcome.

But not so fast. If you read the fine print (or really just the abstract) of DeepMind’s Nature article, AlphaGo isn’t a pure neural net at all — it’s a hybrid, melding deep reinforcement learning with one of the foundational techniques of classical AI — tree-search, invented by Minsky’s colleague Claude Shannon a few years before neural networks were ever invented (albeit in more modern form), and part and parcel of much his students’ early work.

What’s more is that AI still hasn’t reached a level of knowledge and reasoning to deal with questions that require multiple contexts. Indeed, a recent test concluded that present AIs can’t beat an 8th grader.

The Allen Institute’s science test includes more than just trivia. It asks that machines understand basic ideas, serving up not only questions like “Which part of the eye does light hit first?” but more complex questions that revolve around concepts like evolutionary adaptation. “Some types of fish live most of their adult lives in salt water but lay their eggs in freshwater,” one question read. “The ability of these fish to survive in these different environments is an example of [what]?”

These were multiple-choice questions—and the machines still couldn’t pass, despite using state-of-the-art techniques, including deep neural nets. “Natural language processing, reasoning, picking up a science textbook and understanding—this presents a host of more difficult challenges,” Etzioni says. “To get these questions right requires a lot more reasoning.”

It’s only a matter of time until the AI teams get from the 8th grade to high school then to the university level.

How does this relate to games though? With smarter AI comes we will get better bots in games and we’ll see that making NPCs will get easier.

Developing a Unified AI Framework

This month Firas Safadi, Raphael Fonteneau, and Damien Ernst published a paper in the International Journal of Computer Games Technology about how we ought to think about AI in games. They argue that we need a unified framework for dealing with AI development and deployment in games.

Their paper, Artificial Intelligence in Video Games: Towards a Unified Framework, is worth a read and will undoubtedly shape how we think about AI in games for years to come. Think about the possibility that game engines will ship with a suite of default AI behaviours that can be easily modified by non-coders.

Here’s the abstract:

With modern video games frequently featuring sophisticated and realistic environments, the need for smart and comprehensive agents that understand the various aspects of complex environments is pressing. Since video game AI is often specifically designed for each game, video game AI tools currently focus on allowing video game developers to quickly and efficiently create specific AI. One issue with this approach is that it does not efficiently exploit the numerous similarities that exist between video games not only of the same genre, but of different genres too, resulting in a difficulty to handle the many aspects of a complex environment independently for each video game. Inspired by the human ability to detect analogies between games and apply similar behavior on a conceptual level, this paper suggests an approach based on the use of a unified conceptual framework to enable the development of conceptual AI which relies on conceptual views and actions to define basic yet reasonable and robust behavior. The approach is illustrated using two video games, Raven and StarCraft: Brood War.

Page 1 of 5

Powered by WordPress & Theme by Anders Norén