risk aversion in intelligence analysis

One of the things I have been doing for Rapleaf is talking to people that gather and analyze intelligence. We're very interested in systems that gather discreet pieces of info and put them together to find patterns and good intelligence.

while talking to intelligence officers (mostly those in the US govt) i found something very interesting: an incredible fear of failure. the fear for a false positive (falsely identifying something) is so incredibly high that these officers err on the side of massive false negatives (in this case, not going out on a limb or being creative).

essentially, they seem highly risk averse ... and there seems to be very good reason to be risk averse as a false positive is often a career ending move in many intelligence services (often yielding to career stagnation or even firing).

somehow ... and i'm not sure how ... we need to reward people for being wrong. and then protect those people. encourage good mistakes (not stupid mistakes -- there is a big difference).

of course, i'm not an intelligence expert ... but the fear amongst analysts is so pervasive that it MUST be a signal of an endemic problem.