Train2Game News WildMeta Integrating AI Bots With Unity & Unreal

Machine learning is often associated with the few studios that can afford dedicated research teams, but now with WildMeta’s technology and its new Unity and Unreal Engine integrations, machine learning-based bots are more accessible than ever.

Following the footsteps of leading machine learning research organisations, WildMeta went out of stealth in January 2021 and unveiled its own framework to build machine learning-based bots capable of playing video games and react naturally to their environment.

Today the team announces integrations with Unity and Unreal Engine as well as compatibility with most engines thanks to its game developer-friendly library that acts as a bridge between game data and WildMeta’s machine learning systems. 

Over the past year, the company’s founders, Amandine Flachs and Dr Alexandre Borghi have been initiating discussions with game studios on machine learning potential and adoption.

While many game developers showed a genuine interest in the technology, most of their questions have been focusing on how it would actually fit into their existing, most often complex, tech stack. WildMeta’s new integrations and C++ library are a direct solution to these questions. 

In practice, the new library doesn’t change WildMeta’s business model:

  • The team still offers its technology as a service to game studios.
  • The team trains and integrates machine learning-based AIs in games.
  • Studios do not need to hire in-house machine learning experts to start adopting the technology.

Using its own API and library will enable the team to reduce the amount of custom development previously required to make its machine learning technology compatible with games and will let it concentrate its efforts on the game AI itself.

Amandine Flachs, CEO and co-founder of WildMeta said:  

“Bots aren’t a bad thing, dumb bots are! Research teams have shown you can train machine learning-bots to beat the best professional players. We don’t want to beat players, instead, we want this technology to help game developers for example with QA, to improve players’ experience, offer worthy opponents that react to your moves instead of following a script, squad team members that actually contribute to your game or training bots that prepare you to known tactics. We have managed to turn research into a production-ready system that fits studios’ existing tech stacks, now our focus is to drive adoption of this technology.”

Since going out of stealth, the team has given a number of presentations in order to demystify machine learning technology for game development.

They will continue to spread the word in the coming months with more public appearances including a technical deep dive from Dr Alexandre Borghi at Develop: Brighton in October 2021.

For more information go to https://www.wildmeta.com/