The Creators of Baldur's Gate 3 Explains Its Use of Machine Learning for Upcoming Project

The developer behind hit titles like Baldur's Gate 3 and Divinity: Original Sin has recently shown its upcoming project, creating significant anticipation within the industry. However, subsequent remarks from the studio's co-founder have introduced a new dimension to the discussion, focusing on the developer's approach toward machine learning.

AI as a Creative Assistant, Not a Substitute

In a latest clarification, Larian's director outlined that the team is using machine learning for certain supporting functions. These encompass developing PowerPoint slides, producing initial visual ideas, and creating placeholder dialogue.

Importantly, Vincke made clear that the end content in the game will be created exclusively by real creatives. "Our team is writing everything ourselves," he said.

Larian is actively expanding our team of storytellers and are currently assembling narrative groups.

Since this area is being particularly referenced — we currently have twenty-three artistic staff and have roles to fill for more artists.

Everything we do is incremental and focused on having people spend additional energy on actual creation.

Any ML tool used well is additive to a developer's routine, not a replacement for their skill.

Addressing Concerns and Clarifying the Vision

The news of employing this technology initially generated backlash among portions of the community. In reaction, Vincke offered further elaboration on social media.

"Our team utilizes AI tools to explore references, in the same way we use the internet and reference books," he stated. "During the conceptual brainstorming phase we use it as a simple sketch for layout which we then swap out with authentic illustrations."

He noted, "Larian brings on creatives for their creative vision, not for their willingness to execute what a AI generates."

Focused Uses for Machine Learning

Vincke had in the past outlined the team's focused strategy to machine learning, categorizing its use into three main areas:

• Streamlining Repetitive Work: Areas like polishing mocap data, dialogue cleanup, and Larian-specific work like retargeting animations.
• Accelerated Iteration: Using technology to speedily create rough versions of gameplay ideas to validate concepts ahead of full production.
• Long-Term Aspirations: Exploring how machine learning could eventually facilitate emergent gameplay, especially in simulating unforeseen permutations in a complex RPG.

He explicitly affirmed that key artistic areas — including visual art — are not areas where the studio is replacing creative talent. On the contrary, Larian is actively hiring in these very fields.

"Our studio is neither releasing a game with AI-generated content, and we are certainly not looking at cutting creatives to substitute them with artificial intelligence," Vincke stated definitively.