Larian Studios Details Its Implementation of Machine Learning for Next Divinity Game

The studio behind acclaimed role-playing games like Baldur's Gate 3 and Divinity: Original Sin has recently unveiled its next major project, sparking a wave of anticipation within the gaming community. However, subsequent comments from the studio's lead designer have brought clarity to the conversation, addressing the studio's philosophy toward machine learning.

AI as a Creative Assistant, Not a Substitute

In a new message, Swen Vincke detailed that the company is using machine learning for particular supporting functions. These include developing pitch decks, producing initial concept art, and drafting placeholder text.

Importantly, Vincke stressed that the final assets in the game will be created entirely by real writers. "We are creating everything in-house," he stated.

Our studio is actively growing our pool of writers and are actively putting together narrative groups.

Given that this area is being particularly called out — we currently have over twenty artistic staff and have positions available for further artists.

All our efforts we do is additive and focused on enabling creatives to spend additional energy on the creative process.

Every machine learning application applied correctly is a boost to a artist's routine, not a replacement for their skill.

Tempering Reactions with Clear Intent

The news of employing this technology at first generated backlash among portions of the fanbase. In response, Vincke offered more detail on public forums.

"Our team utilizes AI tools to research ideas, similar to we use the internet and physical media," he explained. "During the conceptual planning process we use it as a basic framework for composition which we then replace with original illustrations."

He added, "Larian brings on artists for their inherent skill, not for their willingness to replicate what a machine suggests."

Three Pillars of Practical Application

Vincke had previously broken down the team's practical method to machine learning, grouping its use into three main areas:

• Streamlining Repetitive Work: This encompasses motion capture cleaning, voice editing, and technical processes like adapting animations for different models.
• Rapid Prototyping ('White Boxing'): Using systems to speedily create simple mock-ups of gameplay ideas to test concepts prior to full production.
• Future Potential for Gameplay: Investigating how machine learning could in the future facilitate emergent gameplay, particularly in managing player-driven narratives in a vast role-playing world.

He clearly affirmed that central narrative domains — like writing — are are in no way areas where the team is replacing human input. Conversely, Larian is expanding its staff in these exact roles.

"We are neither releasing a game with AI-generated content, and we are certainly not looking at cutting staff to swap them out with artificial intelligence," Vincke stated definitively.