We study the diagonalization in the context of proof
complexity. We prove that at least one of the
following three conjectures is true:

1. There is a boolean function computable in E
that has circuit complexity $2^{\Omega(n)}$.

2. NP is not closed under the complement.

3. There is no ... more >>>

The working conjecture from K'04 that there is a proof complexity generator hard for all
proof systems can be equivalently formulated (for p-time generators) without a reference to proof complexity notions
as follows:
\begin{itemize}
\item There exist a p-time function $g$ extending each input by one bit such that its ... more >>>