The Creators of Baldur's Gate 3 Explains Its Application of Machine Learning for New Divinity

The studio behind acclaimed role-playing games like Baldur's Gate 3 and Divinity: Original Sin recently shown its upcoming project, generating significant hype within the gaming community. However, recent comments from the studio's lead designer have introduced clarity to the discussion, touching on the developer's philosophy toward machine learning.

AI as a Creative Assistant, Not a Substitute

In a recent clarification, Swen Vincke outlined that the team is using machine learning for certain supporting purposes. These encompass developing pitch decks, generating initial artistic references, and creating placeholder dialogue.

Crucially, Vincke emphasized that the end material in the game will be authored exclusively by real creatives. "Our team is developing all the content in-house," he stated.

We are continuously expanding our roster of writers and are currently assembling dedicated writer rooms.

Since concept art is being specifically called out — we presently have twenty-three artistic staff and have positions available for more creatives.

All our efforts we do is additive and designed to letting our team spend more time on the creative process.

Every AI system used well is a boost to a creative team process, not a replacement for their talent.

Responding to Feedback and Defining the Path

The news of employing this technology originally sparked backlash among a segment of the fanbase. In reaction, Vincke issued more clarification on social media.

"At Larian, we employ these tools to research ideas, in the same way we use the internet and physical media," he stated. "During the initial ideation stages we use it as a basic framework for composition which we then substitute with hand-crafted concept art."

He continued, "Larian brings on creatives for their inherent skill, not for their ability to follow what a AI generates."

Focused Uses for Machine Learning

Vincke had previously detailed the team's practical approach to this technology, categorizing its use into three main functions:

• Handling Monotonous Jobs: This encompasses polishing mocap data, voice editing, and pipeline-specific tasks like adapting animations for different models.
• Fast-Tracked Experimentation: Using technology to quickly build basic models of mechanics to test concepts before full production.
• Long-Term Aspirations: Investigating how AI could eventually enhance emergent player agency, specifically in creating dynamic reactions in a detailed game universe.

He specifically noted that key artistic areas — like writing — are not fields where the company is replacing creative input. In fact, Larian is expanding its staff in these precise roles.

"Our studio is not launching a game with any AI components, nor planning on cutting teams to replace them with AI," Vincke summarized.