The Creators of Baldur's Gate 3 Details Its Use of Machine Learning for Next Divinity Game

The studio behind acclaimed role-playing games like Baldur's Gate 3 and Divinity: Original Sin recently teased its upcoming project, creating immense anticipation within the gaming community. However, subsequent comments from the studio's lead designer have added clarity to the narrative, focusing on the studio's approach toward generative artificial intelligence.

A Tool for Ideation, Not Replacement

In a recent clarification, Swen Vincke explained that the company is using generative AI for specific supporting tasks. These involve developing presentation materials, generating initial artistic references, and drafting temporary copy.

Crucially, Vincke made clear that the shipping content in the game will be crafted solely by actual writers. "We are writing every line manually," he said.

We are continuously expanding our team of storytellers and are busily putting together narrative groups.

As visual development is being specifically referenced — we currently have over twenty concept artists and have job openings for additional talent.

All our efforts we do is incremental and focused on having people spend greater focus on actual creation.

Every ML tool implemented properly is supplementary to a creative team process, not a substitute for their craft.

Tempering Reactions with Clear Intent

The admission of AI usage originally generated backlash among some the player base. In reaction, Vincke provided more elaboration on public forums.

"Our team utilizes these tools to gather inspiration, similar to we use search engines and art books," he wrote. "During the initial planning process we use it as a rough outline for layout which we then substitute with authentic concept art."

He added, "We've hired talent for their inherent skill, not for their willingness to execute what a machine suggests."

Key Areas of AI Integration

Vincke had in the past detailed the studio's targeted method to this technology, grouping its use into key areas:

• Handling Monotonous Jobs: This includes polishing mocap data, dialogue cleanup, and pipeline-specific tasks like adjusting assets for various species.
• Accelerated Iteration: Using technology to speedily create simple models of gameplay ideas to experiment with concepts before full development.
• Experimental Frontiers: Investigating how machine learning could eventually create innovative player agency, specifically in creating unforeseen permutations in a complex RPG.

He explicitly stated that core creative areas — like visual art — are are absolutely not areas where the team is replacing creative talent. In fact, Larian is recruiting more in these exact positions.

"Larian is not shipping a game with any AI components, nor looking at trimming down staff to replace them with artificial intelligence," Vincke summarized.