[liberationtech] NSA-GCHQ meeting on Tor (with slides!)

[Maxim Kammerer]

> I wonder what the current state of affairs is, though. The slides
> suggest that the global passive interception infrastructure is not
> suitable for correlation-based deanonymization, so NSA/GCHQ need
> “access to nodes”. But that was 6 years ago.

See also my analysis from last year [1]. Sniffing ~25 selected C-class
networks with Tor relays gives your ~25% end-to-end correlation
capability. Surely NSA would be able to install 25 designated probes
in 6 years. My guess is that they have the capability, but reserve it
for high-profile national security targets (see last slide).

[1] https://lists.torproject.org/pipermail/tor-talk/2012-August/025254.html

-- 
Maxim Kammerer
Liberté Linux: http://dee.su/liberte