crime, and to addressing the difficulties caused in this field by the increasing use
of encryption.
7.31.
The author of the GCHQ 2015-16 end of year performance report, addressing
GCHQ’s coverage of the Islamist threat outside the UK, noted that the use of
CNE had led to the production of “uniquely valuable intelligence” in respect of the
threat.
Conclusion
7.32.
The bulk EI power is unlike all the others, in that (though the dividing line
between bulk and thematic is not always very clear) it has never been used.
7.33.
It is plain however that, as the internal documents abundantly demonstrate, EI is
a fast-developing alternative to bulk interception (albeit one that in GCHQ’s own
jargon is described as “active” rather than “passive”.
7.34.
I also accept that the logic of bulk interception transposes to EI, in that there will
be foreign-focused cases in which there is significant value to be gained for
GCHQ’s operational purposes but in which it will not be possible to make a
sufficiently precise assessment to proceed on the basis of the thematic EI power.
I would also repeat, as noted at 2.52-2.58 above, that the additional constraints
attaching to the use of bulk EI render it in some respects a more palatable tool
than the thematic EI power.
7.35.
A10/1-2 are both examples of targeted thematic EI in respect of which the only
possible alternative, the use of human sources, was unrealistic. It was possible
to envisage situations fairly similar to those of the case studies in which
insufficient information was available to justify a warrant for targeted thematic EI.
The three hypothetical examples in the open Operational Case (A10/3-5) are
also plausible indications of scenarios in which bulk EI could be needed.
7.36.
For all these reasons, I conclude (as, after full consideration, did the Chair of the
ISC: 3.87(c) above) that an operational case for bulk EI has been made out in
principle, and that there are likely to be real-world instances in which no effective
alternative is available. While it is likely to be of use in particular for the recovery
of equipment data, its capacity to recover content may also be of value (7.7
above).
7.37.
But very considerable caution is required, in view of:
(a) the fact that EI can recover data that has never been sent anywhere (7.16
above);
(b) the untried nature of the power;
109