How will they actually test the viability of 'intelligent traffic systems' with no traffic?
In fact, most of those mentioned systems are about the interaction of that technology WITH PEOPLE in an urban environment. Just an empty urban environment doesn't get you much?
The point of this is that it is a controlled environment, where systems can be tested under lab-like conditions before being released into the urban "wild". Test subjects could be easily recruited to drive around in an instructed manner to replicate any given traffic pattern. This would make it much easier to debug a system before it is tested under real-life urban conditions. And given that this is a lab environment, you can also test the effect of a catastrophic system failures without any safety concerns.
It's a bit more complicated than that.
What people are actually paying for when selling virtual information like photographs and novels are the intellectual property rights pertaining to the work. It is unclear what intellectual property rights a third party developer retains. In many cases the developer of the virtual world will retain the intellectual property right (depending on the licensing agreements in force in the game) and the third party developer is merely selling a "presence" in the game. Without the creative rights associated with it, I'd argue the virtual land would indeed be of little value.
And as to your analogies, in most cases people do pay tens of millions of dollars for paintings for the original, physical copy. Reproductions, even masterfully done ones, are only ever worth a fraction of the piece.
In any formula, constants (especially those obtained from handbooks) are to be treated as variables.