You raise an important point. It strikes at the heart of the problem with running a security service, which is different from criminal law and where the police have to gather enough facts to potentially be able to go to trial. And the individual is aware of the charges. The accused can defend himself, and only admissible facts can be brought against him. So, there is a protection mechanism.
The big problem with intelligence and security is there may be an exchange of information which has implications for the individuals concerned, and which amount to a conviction under criminal law. The worse-case scenario is these individuals may be detained indefinitely, or even tortured as in the cases we are reviewing, without any guarantee there will be a procedure through which they will be able to defend themselves.
And that is why it is imperative, given the implications of such a situation, that there be a solid and credible oversight mechanism to prevent security intelligence activities from depriving persons of their rights.