Game AI Programming | GDC | AI Summit

Game AI Summit (M-T)

Take an inside look at key architectures and issues within successful commercial games in the AI Summit, Monday and Tuesday, March 19-20, 2018. Join top game AI programmers for panels and lectures, in addition to conversations, debates, and rants on how game AI can move forward. Organized as a collective effort by the AI Game Programmers Guild, the summit is targeted to intermediate to advanced programmers who want deeper insight, but anyone interested in what AI can offer next generation games will gain invaluable insights.

View all AI Summit sessions

Featured Sessions

All Access, GDC Conference+Summits, or GDC Summits Pass required to attend summit sessions (M-T).

Beyond 'Killzone': Creating New AI Systems for 'Horizon Zero Dawn'
Julian Berteling (Guerrilla Games)
Having established AI tech in your studio is great… until your next game adds demands well over and above what your existing systems were designed to handle. This session describes the changes that Guerilla made to switch from having to support a single human enemy in closed corridor spaces to a game with more than 25 wildly different characters in a large open world. Specifically, the lecture explains how they changed the navigation and animation systems that made the characters in 'Horizon Zero Dawn' move realistically. Additionally, this session will detail how these changes impacted Guerilla's workflow throughout the project.
'Race for the Galaxy': A Neural Network in Production
Theresa Duringer (Temple Gates Games)
Neural networks have long been viewed with skepticism in game AI. However, there are times when their use is not only appropriate, but a powerful and time-saving approach, particularly for small teams with limited resources. This lecture will explain how 'Race for the Galaxy', a digital adaptation of the board game, uses temporal difference learning to power its AI. This knowledge-free system requires no human input to generate training data, which allows it to improve by playing against itself. Through this approach, the Temple Gate Games team was able to dramatically improve the challenge level offered by AI opponents without the significant time investment typical of tuning complex AI.
Knowledge is Power: An Overview of Knowledge Representation in Game AI
Daniel Brewer (Digital Extremes)
Rez Graham (Independent)
Much time is spent on designing how game agents make decisions. However, regardless of architecture, almost decisions require information from the world. In fact, the quality of the decisions (and subsequent behaviors) is often directly related to the quality of the data available to the AI. It's important to provide this data in a form that is efficient and easily interpreted. The session illustrates and explains a variety of techniques for representing and delivering information. This session will cover systems appropriate to action games such as target information and environmental knowledge. Additionally, this talk will address broader techniques that can be applicable to RPGs and simulation games.
Triage on the Front Line: Improving 'Mafia III' AI in a Live Product
Sergio Ocio Barriales (Hangar 13)
Kate Johnson (Hangar 13)
Changing and testing AI is challenging enough. It becomes even more risky immediately before or after launch when even small tweaks have the potential to upset balance, break scripts, or undermine mission integrity. Development teams have to carefully weigh contradicting user feedback to identify where changes will have a positive outcome for the widest field of players. This talk will explain the methods that were crafted at Hangar 13 to prioritize and address feedback on the AI of 'Mafia III', and the rules for triage and testing of changes. This lecture will also show examples of actual problems that were solved during this process.
Winding Road Ahead: Designing Utility AI with Curvature
Mike Lewis (ArenaNet)
Often, the most difficult part of constructing AI systems is enabling designers to craft behaviors in as intuitive manner as possible. This lecture addresses that issue by showing that powerful, engaging tools don't have to be complex. Using the speaker's free, open-sourced Curvature AI design and prototyping tool as a demo point, this session shows how the software allows designers to specify knowledge representation, enables crafting complicated decisions, and even allows them to see behaviors in a virtual sandbox. The focus is on how programmers can design systems that enable designers to understand what their AI is doing and why.

AI Summit Advisors

Aleissia Laidacker
Magic Leap
Aleissia is a game programmer and designer who has been working as a developer for 16 years and developing games at Ubisoft for 10 of them. She was Lead AI on multiple 'Assassin's Creed' games. Her passion for design has pushed her teams to think about why they are developing their games rather than just how they develop them. She empowers her teams to think outside of the box through game jams and experimentation. Recently she has made the jump from video games to mixed reality, joining Magic Leap as Interaction Director for their new San Francisco office. There, her team focuses on finding new and innovative ways to create rich immersive experiences focused on player interaction and narrative.
Dave Mark
Intrinsic Algorithm
Dave is the president and lead designer of Intrinsic Algorithm, an independent game development studio in Omaha, Nebraska. He does consulting on AI, game design, and mathematical modeling for clients ranging from small indie game studios to AAA companies including EA, Sony Online Entertainment, and ArenaNet.
Dave is the author of the book "Behavioral Mathematics for Game AI" and is a contributor to the "AI Game Programming Wisdom", "Game Programming Gems", and "Game AI Pro" book series. He has also spoken at numerous game conferences and universities around the world on the subjects of AI, game theory, and psychology of games.
He is a founding member of the AI Game Programmers Guild and has been a co-advisor for the previous AI Summits at GDC.
Dave continues to further his education by attending the University of Life. He has no plans to graduate any time soon.
Steve Rabin
DigiPen Institute of Technology
Steve has been a principal figure in the game AI community for nearly two decades. After working as an AI engineer at several Seattle startups, he managed and edited seven game AI books including the recent "Game AI Pro" series and the "AI Game Programming Wisdom" series. He also edited the book "Introduction to Game Development" and has over two dozen articles published in the "Game Programming Gems" series. He's been an invited keynote speaker at several AI conferences, founded the AI Game Programmers Guild in 2008, and founded the GDC AI Summit in 2009. Steve has taught game AI since 2006 at the DigiPen Institute of Technology and has over a dozen issued patents covering profiler analysis, augmented reality, and gesture recognition. He earned a B.S. in computer engineering and an M.S. in computer science, both from the University of Washington.

Follow Us

Our Community

  • Game Developers Choice Awards
  • Independent Games Festival
  • Gamasutra
  • GDC Vault
  • XRDC
  • VRDC