The Creators of Baldur's Gate 3 Explains Its Use of Machine Learning for Next Divinity Game
The team behind popular titles like Baldur's Gate 3 and Divinity: Original Sin has recently shown its next major project, sparking immense excitement within the player base. However, subsequent comments from the company's lead designer have brought a new dimension to the narrative, focusing on the developer's approach toward generative artificial intelligence.
Augmenting Workflows, Not Cutting Jobs
In a latest message, Swen Vincke explained that the developer is utilizing generative AI for particular preliminary purposes. These include fleshing out presentation materials, producing initial concept art, and drafting temporary copy.
Importantly, Vincke made clear that the shipping content in the game will be authored exclusively by actual artists. "We are developing all the content ourselves," he said.
Larian is actively expanding our team of concept artists and are actively forming narrative groups.
As visual development is being explicitly referenced ā we presently have 23 concept artists and have job openings for more artists.
All our efforts we do is supplementary and focused on enabling creatives to spend more time on the creative process.
Every ML tool implemented properly is additive to a creative team routine, never a stand-in for their talent.
Responding to Feedback and Defining the Path
The admission of using AI at first sparked backlash among portions of the player base. In reply, Vincke offered further clarification on social media.
"We use machine learning to explore references, just like we use Google and art books," he wrote. "During the very early ideation stages we use it as a simple sketch for layout which we then substitute with original illustrations."
He noted, "Our studio recruits creatives for their inherent skill, not for their ability to replicate what a machine suggests."
Three Pillars of Practical Application
Vincke had previously detailed the company's practical approach to this technology, defining its use into key functions:
- Automation of Tedious Tasks: Areas like refining animations, dialogue cleanup, and Larian-specific work like adapting animations for different models.
- Accelerated Iteration: Using tools to speedily create simple models of gameplay ideas to experiment with concepts before full development.
- Future Potential for Gameplay: Exploring how machine learning could one day facilitate new forms of reactivity, especially in managing dynamic reactions in a detailed game universe.
He explicitly stated that key artistic domains ā such as writing ā are are in no way fields where the company is reducing human input. Conversely, Larian is actively hiring in these exact roles.
"We are not releasing a game with machine-made assets, nor considering trimming down creatives to replace them with artificial intelligence," Vincke stated definitively.
