I’d like to propose the addition of a Python module for developing and training AI bots in Fortnite, similar to how the RLBot framework works for Rocket League bots. RLBot allows developers to create custom bots in Python and other languages, run matches, and even train them using custom simulations and tools like RLGym, with configurable matches and training setups.
The idea is to have a small training map in Fortnite where bots can be trained in different scenarios (1v1, 1v2, 1v3, 1v4, 2v2, etc.) and where the strongest bots keep learning or can be trained using reinforcement learning or other ML methods. Developers could use a Python API to script behaviors (movement, building, editing, shooting, decision-making) and train bots in simulated matches before exporting them into UEFN or custom matches. Once trained/configured, players and creators could add these bots into UEFN games or custom lobbies with configurable difficulty and behavior.
This would open up AI training and bot creation in Fortnite to the community, making it easier to experiment with advanced bot behaviors like learned building/editing/shooting, similar to the way RLBot makes Rocket League bot development and training accessible.