The Divinity Developer Explains Its Use of Machine Learning for Next Project

The team behind acclaimed role-playing games like Baldur's Gate 3 and Divinity: Original Sin recently shown its new project, creating significant hype within the industry. However, recent statements from the company's figurehead have brought clarity to the conversation, touching on the team's philosophy toward machine learning.

AI as a Creative Assistant, Not a Substitute

In a recent message, Swen Vincke detailed that the developer is employing AI technology for certain preliminary functions. These encompass enhancing pitch decks, generating early-stage concept art, and creating temporary dialogue.

Notably, Vincke emphasized that the end material in the game will be authored exclusively by real writers. "Larian is writing every line in-house," he stated.

We are continuously growing our pool of concept artists and are busily forming writing teams.

Given that visual development is being explicitly referenced — we presently have 23 concept artists and have roles to fill for further creatives.

All our efforts we do is incremental and focused on having people spend greater focus on the creative process.

Every machine learning application implemented properly is supplementary to a artist's routine, never a stand-in for their craft.

Addressing Concerns and Clarifying the Vision

The admission of using AI initially sparked unease among a segment of the fanbase. In reply, Vincke issued further clarification on online platforms.

"We use AI tools to explore references, in the same way we use Google and physical media," he stated. "In the initial brainstorming phase we use it as a basic framework for structure which we then replace with hand-crafted concept art."

He continued, "We've hired artists for their unique talent, not for their capacity to follow what a machine suggests."

Key Areas of AI Integration

Vincke had previously broken down the studio's focused method to machine learning, grouping its use into three main pillars:

• Automation of Tedious Tasks: This includes polishing mocap data, voice editing, and pipeline-specific tasks like adjusting assets for various species.
• Rapid Prototyping ('White Boxing'): Using tools to quickly build basic versions of mechanics to experiment with concepts prior to complete development.
• Experimental Frontiers: Researching how machine learning could one day create new forms of player agency, specifically in creating dynamic reactions in a complex RPG.

He explicitly stated that key artistic disciplines — including music composition — are are absolutely not areas where the company is reducing human talent. Conversely, Larian is actively hiring in these very positions.

"Larian is neither shipping a game with machine-made assets, and we are certainly not planning on cutting staff to replace them with artificial intelligence," Vincke summarized.