“The task is … not so much to see what no one has yet seen; but to think what nobody has yet thought, about that which everybody sees.”
― Erwin Schrödinger

During my decades as an RPG designer, I encountered the issue depicted above at several occasions. I started by writing for Traveller, a gritty space-opera game that resembles the stories that Jerry Pournelle and Poul Anderson wrote in the 1960s and 1970s. In the 1990s, as the IT evolution accelerated swiftly, it became obvious that these writers were unaware of Moore’s Law about the exponential growth of computing power. 1990s computers outperformed their fictional far-future counterparts by several orders of magnitude. An obvious case of “stuff you don’t know that you don’t know”.
By itself, this is a trivial observation that SF fans have known for decades. However, when you write RPGs, you must wrestle with this matter in a different way. A story is “then and there”, set in stone by its author. But a game is never a fixed text, but rather a never-ending creative process by its myriads of players.
My approach: An RPG is a toolbox that the gamemaster uses in any way she sees fit to create her own adventures. Therefore, the tools in the toolbox must make sense. “The computer of this far-future starships uses a mainframe computer that is less capable than my cellphone.” No, that won’t do.
Any game milieu is inevitably deficient when it comes to “stuff you don’t know that you don’t know”.
My academic background is in political science and history. I have learned my limits when it comes to technology and the physical sciences: they’ll always surprise me, that is, “stuff that I know that I don’t know”. (“Methane on Mars!” Wow. “The universe’s expansion rate is increasing!” I didn’t see that one coming. And so on.)
In my own field I end up with: “stuff you don’t know that you don’t know”. For example, I didn’t expect Putin to launch a full-scale invasion of Ukraine last year. At that time, I didn’t know Putin’s reasoning behind his unwise decision (I still don’t). Also, I didn’t see that hole in my knowledge, despite having spent a lot of time studying Russia/USSR at an academic level.
Lessons learned:
A. The past is easier to investigate than the present and the future is out of bounds for any intrepid researcher.
B. If you want to create a credible and durable game, you should make its setting resistant to “ignorance shocks”.
Conclusions:
C. Fantasy milieus operate according to you own set of physical and historical laws. What you know is what there is to know. There might be “stuff that you know that you don’t know” because you have not paid attention to them yet. But you won’t be caught by a “Fall of the Berlin Wall”-style surprise.
D. Recent history with embellishments are also a safe bet. It is easy to research (except for places like North Korea). You can also spice the setting with “classified matters”, such as espionage, conspiracies and weird science, that you design and control. However, tread with care because occasionally old secrets might surface and upset the structure of your setting. Example: The fall of the Berlin Wall opened the East German archives. They keep on revealing new unsavory facts. Your in-game “secret Stasi operations” might therefore turn out to be very different from the schemes those people carried out in reality.