Fenris Creations has open-sourced the Carbon engine framework that powers Eve Online, and the game was cited at IVS2026 as a benchmark for AI agency research.
Fenris Creations released the Carbon engine framework as open source on GitHub in July 2026. The engine was developed and refined over more than 20 years and supports Eve Online's single-shard world. The repository contains over 20 modules covering physics, graphics, networking, and other systems. The engine handled a Guinness World Record battle involving 8,825 players.
Earlier that same week, at the IVS2026 conference, a panel of investors and entrepreneurs from gaming, healthcare, and venture capital discussed AI agents. The session framed Eve Online as the ultimate benchmark for AI agency, noting the lineage from AlphaGo to potential AI in the game. The panel also acknowledged a gap between sophisticated bots and LLM-based agents, and the challenge of making agents economically viable in live games.
The two stories establish Eve Online as both a technical achievement in persistent world infrastructure and a reference point for AI research. The open-sourcing of the Carbon engine invites external developers to study and build on the same technology that supported the record-breaking battle, while the IVS discussion positions the game as a testbed for future AI agents, even as practical deployment remains distant.
Synthesized by Yomimono from the cited Yomimono stories below, each itself sourced, then editorially reviewed. Every fact links the story it came from.
Structured graph also available as JSON at /public/entities/eve-online.
CC BY 4.0.
Fenris Creations has released its proprietary Carbon engine framework as open source on GitHub. The engine, developed and refined for over 20 years, supported Eve Online's single-shard world and its record-breaking 8,825-player battle. The repository includes over 20 modules covering physics, graphics, networking, and more.
At IVS2026, a panel of investors and entrepreneurs from gaming, healthcare, and venture capital discussed the current state and future of AI agents. Topics included the lineage from AlphaGo to potential AI in EVE Online, the gap between sophisticated bots and LLM-based agents, and the challenge of making agents economically viable in live games.