There has been a bit of noise recently with the release of the Stellaris: The Machine Age DLC, as Paradox Interactive released a statement about the AI on Steam which the developers have now attempted to give more information about.
On Steam there is a notice if you scroll down far enough that says:
We use generative AI technologies when creating certain assets. This usually involves ideation of content and visual reference material. These elements represent a minor component of overall development. The AI was used to generate voices for an AI antagonist and a player advisor.
They have to, since Valve I now have new rules about developers stating on Steam pages why they used AI in games.
So the game is clear East using AI content, especially for generated voices. In a way, an AI voice for an AI seems kind of relevant for a sci-fi space game like this, but AI makes a lot of people uncomfortable (and for good reason). This caused some agitation within the Steam Forum with lots of answers. Stellaris game director Stephen Muray responded to Steam’s post to note:
On that note: The AI voice generation tools we use on Stellaris ensure that the voice actors who signed up and built the models receive royalties for every line we create. The ethical use of AI technology is very important to us: we’re pretty good at exploring dystopian science fiction and we don’t want to end up in it ourselves.
I’ll have the team write a dev diary on how we use AI tools in a few weeks.
And then explained further Reddit Also:
We didn’t use it for concept art in The Machine Age – we have some awesome concept artists on our team for that. (You’ll be able to see more of their work in next week’s dev diary.) There may be a few AI-generated pieces on the visdev exploration/mood board, but they would be part of a bunch of other inspiring thematic images.
Personally, I use imaging tools to create basic sketches of things that system designers and I are thinking about, because I’m terrible at art, but I’m pretty good at getting computers to do what we think. (Creating tokens for 4,000 Pathfinder characters that I’ll never play paid off!) The artists then take our ideas and may or may not use them as inspiration to create the final assets. None of these design images enter the game.
We used text-generating AIs for “content ideation”, as we said – basically, content designers can break writer’s block by asking an AI “hey, what are the 40 different things I can find in a mystery box” and see if any of them spark any inspiration. None of the results or generated text enters the game.
We have strict guidelines in place for how we can use AI tools legally and ethically and that we follow.
One of their concept artists also followed say:
I came here late, but for me (not speaking on behalf of the entire art team here), there have been several explorations that leveraged generators, but none made it past the visual development stage, and none was continued in the final artwork.
For me, at least 0% of game assets include any sort of AI.
More importantly: what do you think about this? Let me know in the comments. Currently, the DLC has a very positive rating on Steam.
Article taken from GamingOnLinux.com.
