GDC 2026 AI Executive Panel: Agentic AI in Gaming
The gaming industry continues to navigate change. For years, developers used player actions to operate non-player characters (NPCs) predictably down branched paths with fixed dialogue. Now, agentic AI is flipping the script. Instead of simply following pre-programmed rules, agentic AI operates with autonomy. It acts as an active participant in both the creation of games and the experiences players have inside them.
During a recent panel at Game Developers Conference (GDC), industry leaders from Xbox, Sega, PlayStation Studios and EPAM gathered to discuss this shift. They discussed how autonomous agents change the way teams build games, the ethical challenges of personalized content and why human creativity remains the most important part of the equation.
If you want to understand where the gaming industry goes next, you need to understand agentic AI. Here are the core takeaways from the GDC panel and what they mean for the future of interactive entertainment.
The Evolution of AI in Game Development
Creating a video game requires massive amounts of time, money and highly specialized technical skill. Historically, building a new feature meant writing every line of code by hand, testing it and fixing the inevitable bugs. The introduction of large language models (LLMs) gave developers a helpful copilot for writing code, but agentic AI takes this process significantly further.
Haiyan Zhang, GM of AI Innovation and Science at Xbox, perfectly summarized this transition. She noted that the big shift moves away from simple models that help write code. Instead, developers now use "agentic coding tools where you can give them a holistic end-to-end feature and have that agent go through and deliver that feature."
This does not mean that AI pushes developers out of the studio. Instead, it handles the tedious, repetitive tasks that often bog down production cycles. When teams spend less time wrestling with basic architecture, they can focus entirely on innovation. Francis Hart, VP of Data and AI at Sega, emphasized this benefit, explaining that "by abstracting some of the technical skillset required early prototype stage, it allows everybody to really amp up the creativity."
When developers can prototype faster and iterate more freely, games become deeper and more engaging. As Francis pointed out, "the products that we're delivering to consumers will be much, much richer. It'll be able to have a lot more features."
Infinite Possibilities and Living Worlds
One of the most exciting prospects of agentic AI is its ability to change the game world in real-time. Traditional games offer branching narratives, but those branches are strictly defined by writers years before the game ships. Agentic AI removes those boundaries and allows games to adapt directly to the person holding the controller.
Vitalii Vashchuk, Head of Gaming Solutions at EPAM, painted a vivid picture of this future. He described a scenario in which you are playing your favorite game, and suddenly, the environment starts adapting to you. "So not just to your skill level, but also to your personality, to your choice, even to your mood," he explained.
This level of responsiveness requires a system that remembers your past actions and understands context. Ryan Valenza, Sr. Director of Data & AI at Playstation Studios, highlighted how agentic AI finally makes this possible. He spoke about using AI to create "living, breathing worlds." According to him, AI possesses context. For the first time, developers have a tool that can take a player's exact behavior history and mold the game around it.
This shift moves AI away from simple background text generation. As Haiyan observed, "There's a shift from AI as dialogue to AI as like action and game director." The game itself becomes an active narrator, guiding the experience based on your actions. Valenza summarized the scale of this breakthrough clearly: "What AI unlocks for us...is in the infinite, right? It's in infinite generations of content."
Balancing AI with Human Craft
With AI capable of writing code and generating infinite content, you might wonder if traditional game design will become obsolete. The panelists firmly rejected this idea. While AI provides powerful new capabilities, it lacks the human element that makes a game truly memorable.
Creating a great game requires more than just functional mechanics. It requires taste, pacing, emotional resonance, and a distinct creative vision. Valenza warned that relying solely on technology will never yield a masterpiece. "We believe that a person without taste or craft using these tools will never reach the bar that someone on the creative side can reach," he said.
Instead of replacing the artist, agentic AI acts as a new kind of paintbrush. It empowers creators to build things they could never achieve on their own. Haiyan echoed this sentiment, stating, "I really believe that the AI tools, they're there to help individuals and teams express their own creative voice."
Furthermore, agentic AI does not eliminate the need for experienced software engineers. Valenza noted that while the technology makes it easier to build simpler systems and simpler games, it "does not reduce the need for specialized knowledge in game development." Complex, high-fidelity AAA games will always require deep technical expertise to optimize performance and ensure a smooth experience.
Testing, Safety and the Player Experience
Integrating autonomous systems into live games introduces significant new risks. When a game generates content on the fly, developers cannot manually review every line of dialogue or every new quest before the player sees it. This requires a completely new approach to quality assurance and game testing.
If an AI system hallucinates or generates inappropriate content, it can interrupt immersion and damage the studio's reputation. Valenza offered a practical framework for handling this challenge. "You need to treat player-facing generative AI like an online service," he advised. Just as a studio would load-test a multiplayer server to ensure it does not crash on launch day, they must rigorously "load test generations that are going to reach your players."
The stakes are high when it comes to personalization. Delivering the right content at the right time feels magical but getting it wrong breaks the game’s core loop. Francis warned about the delicate balance of AI curation. "If you give relevant, suggested content and experiences to consumers, then it's great," he said. "If you're doing the opposite, it's almost toxic to the experience."
A New Genre: The Rise of the AI Game
As agentic AI becomes more accessible, it will inevitably give rise to entirely new genres of interactive entertainment. Some players will specifically seek out games driven entirely by AI, eager to see how the systems react and evolve.
Francis highlighted this emerging demographic, noting, "There's a market now for people who want to buy an AI game... You look at the players and what they want from those games is the idiosyncrasies of AI and how it behaves, right?"
Interestingly, this hyper-advanced technology might also spark a nostalgic revival of classic gaming formats. Because AI excels at text generation, we may see a resurgence of text-based adventures and Multi-User Dungeons (MUDs). Haiyan playfully noted this trend, "ASCII art is back." She recalled her advice from the previous year's GDC panel: "Hey, we should all be making MUDs. We should all be making text-based games."
Whether through text-based adventures or fully immersive 3D worlds, the goal remains the same: giving players more freedom. As Haiyan pointed out, "AI is there to open up more options for the player." Ultimately, agentic AI delivers a paradox. By using complex, automated systems, developers can offer a "more curated experience that allows you more agency," as Valenza described it.
The Next Level of Game Design
Agentic AI marks a fundamental shift in how we create and play video games. It moves the industry away from static, pre-programmed environments and toward dynamic, infinite worlds. It empowers developers to prototype faster and enables small teams to execute massive ideas.
However, technology alone cannot make a great game. The human element remains the most critical component of game development. As studios learn to balance the raw power of agentic AI with meticulous human oversight, players can look forward to experiences that are deeper, richer and more personalized than ever before.
If you are a developer, now is the time to start experimenting with agentic tools in your prototyping workflows. If you are a player, get ready. The games you play are about to start playing along with you.
