---

WORKSHOP ON AUTOMATED GAME DESIGN (WAGD)

---

The goal of the Workshop on Automated Game Design (WAGD) is to provide a home for advancing the conversation and state-of-the-art in automated game design (AGD).

WHY AGD NOW?

Game design is increasingly becoming malleable, something that players engage with themselves, something that is procedurally generated and varied, and something that is fluid and always changing. The popularity of game-making tools like Fortnite Creative, Mario Maker, as well as the existence of popular ROM-hacking and randomizer communities has encouraged an influx of people to play with game-making, which makes the study and understanding of game design and its underlying processes more important than ever. Daily challenges and roguelike variation now often reconfigure fundamental aspects of a game’s design as it is played, making it important to build better tools to understand and predict dynamic systems with an incomprehensible number of potential outcomes. The ubiquity of live access, games-as-a-service and early access models means that even without runtime modifications, game designs are constantly in flux, which creates an opportunity for researchers to study how intelligent systems and tools can understand and support designers working in these rapidly-changing game design spaces.

Beyond this, AGD is a forward-looking, exciting blue-sky research space that aims to open up entirely new ways to play and create games. AGD carves a unique space for designers and researchers to explore how we build systems that produce surprising and diverse game artifacts. AGD lets us explore the tooling at design- and run-time that pushes the envelope of what styles of game design and player engagement are possible. AIIDE is an ideal venue for such a workshop: it has historically been a home for AGD research, forming a conversation between maximalist approaches to automated game design and more specific game AI, mixed-initiative tooling, and procedural content generation research. Hosting this workshop can help turn up the volume of this conversation.

GET INVOLVED

We are now accepting submissions for publication and presentation! Interested in joining, but not interested in submitting a paper? You are still welcome! Between paper presentations and collaborative design sessions, there is still a lot to offer. If you would like to attend, please register at the AIIDE 2026 website. In person attendance is required for workshop participation.

Please submit any questions to samuel.m.shields@gmail.com.

---

SCHEDULE

---

We plan on having three major sections to the workshop over two days (Monday, November 9th, and Tuesday, November 10th):

• Keynote Address
• Paper Presentations
• Collaborative Demo/Jam Session

Stay tuned for specifics closer to the workshop!

---

CALL FOR PAPERS

---

Submitting Your Work:

We plan on having accepted peer-reviewed papers published in CEUR-WS. As such, submissions follow the length guidelines of CEUR-WS (5-9 pages for short papers, 10 or more for full, references and appendices included) and can be anything related to the field of AGD. This includes theory and vision papers, lofi/hifi prototype descriptions, and evaluations of AGD systems. We encourage maximalist approaches; AGD systems often sum many disparate pieces together. Don’t be afraid to be messy or over-scoped. Papers focused on large pretrained generative models are discouraged due to cost barriers, replicability concerns, and sufficient existing coverage at other venues. We will review artifact evaluations alongside traditional papers.

LaTeX, Overleaf, and other templates for authoring can be found below. All papers must use one of the following templates:

• CEURART.zip
• Overleaf Template

All submissions will be peer reviewed. Works will be judged on their technical and research contributions, accessibility, replicability, and impact. All accepted works will be published in the CEUR-WS proceedings for WAGD 2026. Submissions should be made through Easychair.

Submissions can be made at Easychair (Automated Game Design 2026 track).

Interested in coming to the workshop but don’t have a submission? Want feedback on a demo, or want to ideate on a new system? Register anyways! All are welcome to come and learn more about AGD.

Examples of Relevant Topics:

Below is a (non-exhaustive) list of topics we are looking for:

• Languages and formalisms for describing game design spaces. See VGDL (Ebner et al., 2013) or Cygnus (Summerville et al., 2017).
• Automated systems that design physical or digital games. See Ludi (Browne, 2008) or ANGELINA (Cook, Colton and Gow, 2016).
• Approaches for evaluating unseen games or spaces of games. Includes evaluating large design spaces for playability, balance, or emergence. See Expressive Range Analysis (Smith and Whitehead, 2010).
• Mixed-initiative approaches to automated game design. See Tanagra (Smith, Whitehead, and Mateas, 2010) or Sentient Sketchbook (Liapis, Yannakakis, and Togelius, 2013).
• Studies or surveys of automated game design or similar works in commercial games. See Mosa Lina (Steele, 2023) or Caves of Qud (Freehold Games, 2024) as example games to analyze.
• Novel methods or combinations of methods for automated game design. See WaveFunctionCollapse (Karth and Smith, 2017).
• Automated generation of game mechanics, rules, or progression systems. See Zook and Riedl’s work on RPG and Platformer games (Zook and Riedl, 2014)
• Self-modifying game systems that change over time, either through engagement with players, a larger metagame, or story progression. See Galactic Arms Race (Hastings, Guha, and Stanley, 2009).
• Systems that generate games scoped within a single platform or genre. See Karth et. al’s work on Game Boy RPGs (Karth et al., 2021), or BrawlerAGD (Shields et al., 2022).
• Approaches to the orchestration of multiple distinct creative facets or modalities. See Liapis et al.’s work defining orchestration as a central challenge for AGD research (Liapis et. al, 2019).
• Code-level AGD: synthesizing human-readable and human-editable code (within the context of an existing codebase) as an AGD approach. See Butler et al.’s work on program synthesis for boss behavior generation (Butler, Siu, and Zook, 2017) and Cook’s work on AGD-friendly software design patterns (Cook, 2020).
• Techniques for inventing good game goals. See Davidson et al.’s work on searching for human-like goals in open-ended play environments (Davidson et al., 2025).

Key Dates:

All deadlines are at 11:59 AOE (anywhere on Earth).

• Submissions Open: June 18th, 2026
• Submission Deadline: September 18th, 2026
• Review Deadline: September 25th, 2026
• Submission Notifications: October 2nd, 2026
• Publication-Ready Submission Due: October 16th, 2026

Questions on submissions, want guidance, or not sure if you’re a fit? Send an email to samuel.m.shields@gmail.com.

REFERENCES

• Browne, C. B. (2008). Automatic generation and evaluation of recombination games (Doctoral dissertation, Queensland University of Technology). [LINK]
• Butler, E., Siu, K., & Zook, A. (2017, August). Program synthesis as a generative method. In Proceedings of the 12th International Conference on the Foundations of Digital Games (pp. 1-10). [LINK]
• Cook, M. (2020, August). Software engineering for automated game design. In 2020 IEEE Conference on Games (CoG) (pp. 487-494). IEEE. [LINK]
• Cook, Michael, Simon Colton, and Jeremy Gow. "The angelina videogame design system—part i." IEEE Transactions on Computational Intelligence and AI in Games 9.2 (2016): 192-203. [LINK]
• Davidson, G., Todd, G., Togelius, J., Gureckis, T. M., & Lake, B. M. (2025). Goals as reward-producing programs. Nature Machine Intelligence, 7(2), 205-220. [LINK]
• Ebner, M., Levine, J., Lucas, S. M., Schaul, T., Thompson, T., & Togelius, J. (2013). Towards a video game description language. [LINK]
• Freehold Games. (2024). Caves of Qud (Version 1.0) [Video game]. https://www.cavesofqud.com/ [LINK]
• Hastings, E. J., Guha, R. K., & Stanley, K. O. (2009). Automatic content generation in the galactic arms race video game. IEEE Transactions on Computational Intelligence and AI in Games, 1(4), 245-263. [LINK]
• Karth, I., Duplantis, T., Kreminski, M., Kashyap, S., Kukutla, V., Lo, A., ... & Smith, A. M. (2021). Generating playable rpg roms for the game boy. [LINK]
• Karth, I., & Smith, A. M. (2017, August). WaveFunctionCollapse is constraint solving in the wild. In Proceedings of the 12th International Conference on the Foundations of Digital Games (pp. 1-10). [LINK]
• Liapis, A., Yannakakis, G. N., Nelson, M. J., Preuss, M., & Bidarra, R. (2018). Orchestrating game generation. IEEE Transactions on Games, 11(1), 48-68. [LINK]
• Liapis, A., Yannakakis, G. N., & Togelius, J. (2013). Sentient sketchbook: computer-assisted game level authoring. [LINK]
• Smith, G., & Whitehead, J. (2010, June). Analyzing the expressive range of a level generator. In Proceedings of the 2010 workshop on procedural content generation in games (pp. 1-7). [LINK]
• Smith, G., Whitehead, J., & Mateas, M. (2010, June). Tanagra: A mixed-initiative level design tool. In Proceedings of the Fifth International Conference on the Foundations of Digital Games (pp. 209-216). [LINK]
• Steele, L. (2023). Mosa Lina [Video game]. Lawrence Steele. https://store.steampowered.com/app/2116860/Mosa_Lina/ [LINK]
• Summerville, A., Martens, C., Harmon, S., Mateas, M., Osborn, J., Wardrip-Fruin, N. and Jhala, A., 2017. From mechanics to meaning. IEEE Transactions on Games, 11(1), pp.69-78. [LINK]
• Zook, A., & Riedl, M. (2014, June). Automatic game design via mechanic generation. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 28, No. 1). [LINK]

---

COMMITTEE

---