The Creators of Baldur's Gate 3 Details Its Use of Machine Learning for Next Project
The team behind acclaimed titles like Baldur's Gate 3 and Divinity: Original Sin has recently unveiled its new project, sparking immense anticipation within the industry. However, recent comments from the studio's lead designer have added nuance to the narrative, focusing on the team's philosophy toward machine learning.
A Tool for Ideation, Not Replacement
In a recent clarification, the studio's founder explained that the team is employing machine learning for specific ancillary functions. These include developing PowerPoint slides, creating rough visual ideas, and creating placeholder copy.
Importantly, Vincke stressed that the final content in the game will be created solely by human writers. "Our team is creating all the content in-house," he affirmed.
Our studio is constantly expanding our team of writers and are actively forming writing teams.
Given that this area is being explicitly called out — we currently have over twenty concept artists and have roles to fill for additional talent.
Everything we do is additive and designed to enabling creatives to spend more time on the creative process.
Every AI system applied correctly is additive to a creative team routine, never a stand-in for their talent.
Tempering Reactions with Clear Intent
The news of using AI originally generated backlash among portions of the fanbase. In response, Vincke issued more detail on online platforms.
"We use these tools to gather inspiration, similar to we use the internet and physical media," he wrote. "In the very early planning process we use it as a rough outline for structure which we then swap out with original artwork."
He continued, "Larian brings on creatives for their unique talent, not for their willingness to execute what a AI generates."
Three Pillars of Practical Application
Vincke had in the past outlined the studio's practical approach to machine learning, categorizing its use into three main pillars:
- Automation of Tedious Tasks: This includes polishing mocap data, voice editing, and technical processes like adjusting assets for various species.
- Fast-Tracked Experimentation: Using systems to quickly build rough mock-ups of scenarios to experiment with concepts ahead of full implementation.
- Experimental Frontiers: Exploring how machine learning could eventually create emergent player agency, particularly in creating player-driven narratives in a detailed game universe.
He specifically stated that core creative areas — like music composition — are not fields where the team is replacing human involvement. In fact, Larian is recruiting more in these precise positions.
"Our studio is neither launching a game with AI-generated content, nor considering cutting teams to replace them with AI," Vincke stated definitively.