Following recent events, I feel that our understanding of adversarial computational models had greatly improved, and I would like to suggest a few new models:

  1. All knowing adversary with unbounded computational power would be known as very weak NSA adversary.
  2. A very weak NSA adversary equipped with secret courts, and unbounded budget is a weak NSA adversary.
  3. A weak NSA adversary that answers any query with automatic labelling of a traitor is an insecure weak NSA adversary.
  4. An insecure weak NSA adversary with backdoors into your Turing machines is an NSA adversary.
  5. An NSA adversary that wipes your memory every time you think about the NSA is a strong NSA adversary.
  6. A strong NSA adversary that improves the security of anything in any way is a snark NSA adversary.

Comments are closed.