>"an organism that sees reality as it is will never be more fit than an organism of equal complexity that sees none of reality but is just tuned to fitness
Hmm, I'm curious about this. No animal can see high energy gamma radiation and other dangerous radiation sources, but the human can detect them with tools. That means the better detector of reality will always have a higher fitness level (we can avoid Chernobyl's radiation 10,000 years from now after its long forgotten) but animals won't. A lot of good having 10 babies is if they all succumb to radiation poisoning before they can reproduce, or have mutations that make them less fit in other ways. We can extend this to other ways like detecting disease early, seeing microrganinisms, etc. Clearly fitness is tied with some level of reality detection, at least for advanced tool using species like ours.
That said, there's nothing wrong with the idea that we might be a simulation in a larger reality or many simulations down. Penrose may not be getting his point accross well, but that idea is philosophically sound.
I don't think he's any sort of crank. He may be playing up the whole "I can model the universe to show you its really a virtual machine" but that's its own sort of mathematical solipsism imo. He does make many good points about perception, relatively, etc. Its not the usual "hey guise, quantum physics means magic is true" crap you see on the internet.
The problem is we don't know if its even possible to detect if we're being virtualized. The funny thing is that for most/all of humanity's existence this was an assumption; the world was fake or at least a temporary placeholder and the true reality was some type of spiritualism/heavenly realms. Interesting that we're going back to that in a round-about way.
I'm glad someone is taking this hypothesis seriously. Detecting if we are virtualized is just as important, as say, SETI. People used to laugh at SETI. Why are we laughing at this?