This and other similar patents not yet noticed are inevitable.
For defensive reasons at a minimum but also because generative software is already being used on game development. And this djinn is not going to be bottled. And it is being used because it is *needed*
As the oft-ignored truism says: "AI is not going to take your job, but a human using AI will."
Software that amplifies the productivity of a single developer to do the work of several humans *will* be used. If nothing else because game development costs are spiralling out of control at the same time sales of those expensive games are becoming more and more volatile. Games that succeed, rake it in big time, while the others thud badly. No middle ground. It's not necessarily that the less successful games are terrible or even bad at all, it's just that with 9-figure budgets even falling "short of expectations" can be deadly to a studio, its staff, and even the publisher. Because just like Hollywood movies, games are fully funded over years before release, and the market can and often does change in the interim. Midjudge the market's mood and loose the audience. And the investment.
Anything that can reduce the cost of development and/or allow the flexibility to adjust the game as close as possible to the release date (read the patent's prose carefully) or even during gameplay, will be welcome as a tool to minimize risk.
Consider just how much grief UBISOFT would've been spared if AC SHADOWS had a character creator instead of two hard wired defaults. (MASS EFFECT got it right.) Or, similarly if VEILGUARD actually responded to user choices and preferences instead of presenting "choices" that had no impact on the narrative. Again, Bioware *used* to do it right in both MASS EFFECT and DRAGON AGE. The game narrative responded --within limits--to player choices and preferences.
Of course, developing decision trees that extensive takes time, money, and *foresight* to consider a broad range of gamer preferences and the state of the market years and years in advance. The alternative is ending up with a $200M game that struggles to break even instead of doubling or tripling the investment. A failing that few publishers can survive.
So, like it or not, Generative Tools will be used.
Because studios that don't will be literally betting the farm on five or ten year old choices. And end up with a CONCORD instead of a RIVALS.