> He conflates absence of evidence with evidence of absence.
Absence of evidence is evidence of absence. The false statement of that form is "absence of proof is proof of absence".
If absence of evidence wasn't evidence of absence, then, at best, the presence of the event and the presence of evidence are have no bearing on each other & are independent (in the statistical sense, i.e. the "evidence" is not evidence at all), and at worst, presence of evidence corresponds to the event not happening (i.e. the "evidence" is backwards).
Absence of evidence is [pretty much the weakest possible] evidence of absence. It is most useful for things which are readily apparent. It is not particularly useful for determining questions where the evidence is not readily apparent (such as, obviously, 'Is the NSA spying on Americans?').
One may easily use the absence of evidence of woolly mammoths to conclude that woolly mammoths are extinct. Given their physical size (they are easy to detect) they could not remain hidden from view. We can assume (from our knowledge of other large mammals) that they would leave signs of their presence wherever they trod (footprints, dung, etc.) It is not so easy to use absence of evidence to disprove the existence of neutrinos or tree shrews. Both neutrinos and shrews are hard to detect. There are few people even qualified to attempt to detect them, fewer who are interested, and even fewer who have the resources and opportunity.
A relevant question is "Is it possible for spies to conduct espionage undetected for an extended period of time?" The answer is yes, there is ample evidence of that. Schneier is clearly in a place of trust in his community, making him a high value target. The standards of evidence for betrayal of trust being high (and difficult to define), mean that Schneier could potentially be a spy/collaborator in some capacity, and/or could become one at any time.
This is only the case for binary events (either "it happened" or "it didn't"), which in the real world are quite rare. "Absence of evidence" usually means "The test hasn't been run yet", or "The test results are inconclusive." This is, of course, different from "The test has been run, and I'm choosing to rationalize the results after the fact."
Furthermore, I took it as him saying that he could provide a list of reasons that people should not trust him but he won't because it would be the security pundit version of cock-blocking himself.
No, he is suggesting "innocent until proven guilty", you are suggesting the opposite.
This is a "have you stopped beating your wife" question, what could any person say except "there is no evidence that I am untrustworthy". How do you mathematically verify a person?
The technically correct answer is "mu" or "wu". It's an asian term for absence. In the context of a questions, it means that the question cannot be answered correctly due to a lack of preconditions or because the answers are restricted to a set which does not contain the correct answers.
"Did you stop beating your wife" must be answered wu if you have not been beating your wife. "stopping X" requires the precondition that X was happening at the time of the question.
"Am I trustworthy", as the parent comment pointed out, must be answered wu as well, because it probably is impossible to be universally and unconditionally trustworthy.
Even if it is not, judging if one will be unconditionally trustworthy in all contextes, actions and all possible futures of this situation cannot be estimated with enough confidence, unless your life will end within a very short timespan.
The better question to ask would be "Why might I be trustworthy" which he could then try to answer. Not that I don't think Schneier is untrustworthy..