I was thinking recently about a few things:
- The common acceptance of "you can't prove a negative" in academic papers.
- The existence of "no signalling" theorems (attempts to prove negatives.)
- Entanglement doesn't work for signalling because observing the photon for data changes the photon, and currently they have to send the original measurements for comparison.
- Recent quantum/gravitational wave research involves "partially observing" subjects to avoid changing them.