Generative AI: Expanding the Boundaries of Creativity in Gaming with Hillary Mason

In a recent episode of ODSC’s Ai X Podcast, Hillary Mason, a prominent data scientist and co-founder of Hidden Door, delved into the intersection of generative AI and video games. Throughout the conversation, Mason offered insightful perspectives on how generative AI is reshaping creative fields, particularly gaming, and highlighted both its promise and challenges. This blog post captures the key takeaways from her discussion, focusing on generative AI’s role in gaming, creative storytelling, and the future of AI in building immersive worlds.

You can listen to the full podcast on Spotify, iTunes, and SoundCloud.

Generative AI: Expanding the Boundaries of Creativity in Gaming

Hillary Mason’s work with Hidden Door, an AI-driven storytelling game studio, centers around using generative AI to transform the ways players engage with fictional worlds. According to Mason, generative AI opens up new possibilities that were previously unachievable in digital media. One of the studio’s primary goals is to create story-driven games that go beyond the traditional linear narratives found in many modern games. Instead, they aim to craft experiences where the world reacts and evolves based on the player’s choices — an ambition made possible by generative AI.

Hidden Door’s core concept focuses on allowing players to interact with stories in ways that mirror how they engage with books or movies. Imagine reading a novel and then having the opportunity to experience it as a dynamic role-playing game, guided by AI that shapes the narrative based on the player’s decisions. For Mason, generative AI is the key to unlocking this sort of interactive fiction — where every player’s experience is unique and deeply personalized.

However, while the potential for generative AI to enhance creativity in gaming is undeniable, Mason is careful to clarify its limits. “Generative models are not creative,” she emphasizes. Instead, they are tools for augmenting human creativity. They can help build immersive environments, but they are fundamentally pattern-matching machines, designed to produce an output that approximates the average of the data they’ve been trained on. Thus, the real creativity in AI-driven games comes from the humans behind the scenes — writers, designers, and developers — who use AI to extend their vision, not replace it.

Generative AI in Gaming: Storytelling Meets Interactivity

One of the most exciting applications of generative AI in gaming, according to Mason, is the creation of Non-Player Characters (NPCs) and dynamic game environments that respond believably to the player’s actions. Mason references earlier gaming experiments that used AI for NPC interaction, but notes that these systems often failed to deliver compelling experiences. Many early implementations allowed players to ask NPCs questions or engage in conversations, but the interactions were shallow and often broke the game’s immersion.

Hidden Door is taking a different approach. Instead of focusing solely on making NPCs that players can chat with, Mason’s team is working to make the entire world of the game feel alive and responsive. For example, if a player chooses to burn down a city, that action has lasting consequences, and the world changes in response to it. Hidden Door allows players to type in commands that shape the story, and if the player’s actions align with the rules of the world, the AI dynamically generates responses. This approach is designed to create a more fluid and engaging experience, where players feel that their choices genuinely matter.

The Challenges of AI in Creative Domains

Despite the promise of generative AI in gaming, Mason also acknowledges its significant challenges. One of the most pressing issues is immersion. She points out that systems relying on large language models (LLMs) often face technical limitations, such as delays in response time, which disrupt the player’s experience. Additionally, allowing unrestricted conversations with AI-driven characters often leads to incongruent interactions that break the flow of the game.

Mason emphasizes that for generative AI to create truly compelling gaming experiences, developers need to move beyond simple chat interfaces. Chat systems are ideal for text-based games or brief interactions, but they often fall short in complex, immersive environments. Instead, the focus should be on integrating AI into the game world in ways that feel natural to players and enhance their sense of being part of a living, breathing universe.

This leads to another of Mason’s core arguments: generative AI excels at producing average results based on the data it’s trained on, but it lacks the ability to generate truly original or inspired content. As a result, the role of the human creator remains central in building the scaffolding around which AI can generate content. At Hidden Door, this philosophy manifests in the use of AI to augment human-designed worlds, characters, and stories, rather than attempt to replace them.

Ethical Considerations and AI in Gaming

Mason is well-known for her commitment to ethical AI, and this topic was a recurring theme in the podcast. In the gaming industry, where generative AI has the potential to replace human labor (such as artists and writers), ethical concerns abound. Mason cautions against using AI merely to cut costs by reducing human involvement, arguing that this approach often leads to inferior products and misses the larger opportunity of using AI to enhance human creativity.

At Hidden Door, Mason and her team have implemented policies that prioritize ethical practices, such as hiring human artists and ensuring that their work is incorporated into the game’s AI-generated environments. Rather than using AI to replace artists, they use it to complement and enhance their work. This ensures that the creative vision behind the game remains intact, while still benefiting from the efficiencies AI can provide.

Mason also touches on the importance of data transparency and AI ethics in game development. With generative AI models, the quality of the output depends heavily on the data used to train them. She stresses that it’s crucial for companies to be transparent about their data sources and ensure they are using data ethically. This includes paying creators for their contributions and giving them control over how their work is used. The notion of “artisanal Brooklyn AI,” as Mason playfully refers to it, underlines Hidden Door’s commitment to blending handcrafted content with AI, ensuring that the technology enhances rather than exploits creative labor.

Conclusion: The Future of Generative AI in Gaming

Hillary Mason’s conversation on the AIx Podcast offers a forward-looking yet pragmatic perspective on the future of generative AI in gaming. While the technology has the potential to revolutionize how we create and experience games, it is not without its challenges. At its core, generative AI is a tool that can augment human creativity but not replace it.

For Mason, the future of gaming lies in leveraging AI to build more immersive, dynamic, and personalized experiences, where players’ choices matter and the game world reacts in believable ways. As AI continues to evolve, game developers will need to find innovative ways to integrate the technology without losing sight of the creative vision that makes their games unique.

By focusing on responsible AI practices, ethical data usage, and a commitment to human creativity, studios like Hidden Door are paving the way for the next generation of AI-driven games. The question is not whether AI will play a role in the future of gaming — it already is — but rather how we can ensure it does so in a way that enriches the gaming experience for both creators and players alike.

Originally posted on OpenDataScience.com

Read more data science articles on OpenDataScience.com, including tutorials and guides from beginner to advanced levels! Subscribe to our weekly newsletter here and receive the latest news every Thursday. You can also get data science training on-demand wherever you are with our Ai+ Training platform. Interested in attending an ODSC event? Learn more about our upcoming events here.