Bigger projects take longer, and many studios chased bigger and better, instead of aiming smaller. Which then meant that these things had to be standout hits to be considered successful, which puts more pressure on perfecting everything, which takes more time, ...
Then many studios have stories of either deep direction changes or large projects being cancelled as the money wasn't as available anymore and trends shifted. E.g. Sony alone has probably spent over a billion and multiple years of work of several studios on live-service games that they then scrapped, Dragon Age Veilguard pivoted a few times (and ended up a mess as a result), ... stories like that across the AAA space.
Will Wright said that you go from 4 hours of work per hour of meetings to 1 hour of work per 4 hours of meetings. This can be scaled down with wikis and tools. Chinese companies tend to have "living documentation" aka system analysts - a person whose job is to understand everything and be asked questions whenever someone gets stuck. Western companies are very resistant to this model for some reason.
Bethesda is another way to build things at scale. Lots of isolated sections that don't link into the main thing. But it's hard to build something like Fallout 2 that way when the whole story isn't in one person's head.
Part of the reason for so many layoffs recently is because it works. It's a magic tool that increases speed and reduces costs, and you just pay for it with everyone's motivation and soul.
I've been wondering about this as well, here are some of my hypotheses...
Business related
* Covid did it: we are currently experiencing a delayed effect. AAA games take roughly 5 years to develop (I think?), Covid generated a lot of uncertainty and chaos and thus less games got started during this period. Supply chains and finance were in disarray and we didn't know how long the pandemic would last. To make matters worse, this coincided with the release of the new console generation, which is typically a big moment of reorganizing and new ip, so we had a double whammy in that regard. How much new consoles would get sold in this new environment of chip shortages? How long would the shortages last? Is crypto here to stay and will they keep buying up al GPUs?
* Lending has gotten more expensive and this is driving up risk perception and lowering profitability. In the aggregate this leads to less games being released.
* There's been a lot of mergers between industry behemoths, so internal strategies and reorgs might have something to do with it. There is more vertical integration. There is less competition between AAA publishers because there are simply less of them.
* Outsourcing has changed due to internal and geopolitics. Chinese game market development and government restrictions, etc.
* There's been a switch to mobile and PC gaming, the home console market has been shrinking in relative terms.
* Online game libraries on consoles keep expanding, resulting in lower demand for new games that we saw in previous generations.
Game trends related
* All AAA devs were making their variation on open world games, but there has been an increasing fatigue with the formula. I see something similar happening roughly once every decade where the AAA industry grapples to find new mechanics and a corresponding production pipeline. Like in the 2000-2010 period games became highly linear because that way you could slice the units of work in time. Then in 2010-2020 people got sick of that and the new production pipeline became one of open world freedom alongside editors and workflows that allowed the entire team to populate a single giant world. Gameplay changed from highly scripted cinematic experiences into 'immersive' gameplay systems. Player now are getting tired of this trend but we haven't figured out a new one yet. Before the 2000s this was less of an issue because dev teams were more nimble, asset creation was less of a bottleneck and technical innovations were a bigger driver in popularity and length of development.
* There's been a switch to "forever" games in AAA. Where a studio just builds one big game they keep supporting and funneling players into. These are typically multiplayer, but that leads to issues of network effects. There's been a lot of new multiplayer games over the past decade but the player pool hasn't kept the same pace, so only the popular ones and first movers survive.
* Indie games glut + higher quality indie 'AA' games causing the AAA sector to having to differentiate themselves more, which they suck at because they're all about return on investment.