Bruno Marchal wrote:
For example they will correctly infer some standard model particle theory from they high level experimentations, but as soon they will build particle accelerator to verify their theories, discrepancies will appear (just because we have not simulate the society-world at such a detailed level. So now those researchers can infer that they are simulated at some different reality level. But this is what we don't want. So let us add a subroutine which observes the researchers, and each time reserachers find (serious) discrepancies, the subroutine freezes the researchers and refines their level of reality.
...and there is another theory which says this has already happened... D. Adams.