This is just the games-specific version of the tech industry special-snowflake argument. Building entertainment software at a AAA scale isn't really that different from anything else: you still need a strong plan with good coordination and clear delineation about who can make what decisions to succeed; you still have a group of target customers to impress, and a process of iterative refinement if you don't meet their needs first time out; you're still building something whose primary purpose for existing is to capture customer revenue and whose creative elements are subsumed into that purpose. It's a field with a pretty high level of variance in interesting problems on the micro level but a very heavily circumscribed scope on the macro level outside the few people lucky enough to work on the really innovative stuff.
Like, are you really gonna tell me that the people sitting down to create the latest annual Call of Duty are going in blind to a process dictated purely by the unpredictable artistic muse? That game is getting scoped out tightly to hit a mandatory deadline, its design is getting driven by a brainstorming process for new market-distinguishing features, and it's getting built by experienced craftspeople who are applying tremendous skill to doing something they've trained to be really good at. It's more like launching a new annual fashion line than writing a novel or building conceptual sculpture for a gallery show.
I think your example proves my point:
CoD is the perfect example of production efficiency you are saying the games industry doesn't have. Its a multi-studio machine that churns out product to hard marketing deadlines annually, whilst trickle feeding post-launch content until the next iteration launches and takes up the role.
But as a result the end product is rarely creatively inspired or radically different from its predecessor.
You want better widgets, improve production efficiency. You want better art, give the creatives more time and latitude to experiment and fail if they need to.
The reality is that the best "production plan" in the industry is find something that sells, and ride that pony hard until its legs fall off. The hard part of course is finding that prize pony in the first place, or a successor when the old one gets shipped off to the knacker's yard.
So no, its not special snowflake syndrome, its the reality of chasing hits in a highly competitive marketplace where noone really knows for sure what's going to blow-up huge and what's going to tank.
Meanwhile market expectations are forcing increased head-count and costs without immediate monetary return, with their ongoing accrual making the bean-counter's nervous and encouraging them to impose creative changes to mitigate investment risk.
This stuff is HARD. Its hard technically, its hard creatively, and its hard politically.
What's especially bad in software development is that its not like movies or music which are composites of different discrete elements, simply layered and assembled to create a final product. Games are fundamentally mechanical in nature, you alter one thing and its likely to have unforseen consequences in a dozen other areas or situations. Of course the bigger and more complex the project, the worse it gets - and it all takes time to work through.
And look, the bottom-line is that you can get the best people, give them a ton of time and money, organize them with military precision, market-research and focus-test the shit out of their work, and still come up with a mediocrity or worse.
That's modern Hollywood for you.