







We learned some things this week about the US Predator drone program that has some people appalled and indignant - it is the kind of story that makes news.
A Predator drone is an unmanned aerial vehicle (UAV) used by the United States Air Force both for reconnaissance and for offensive operations. It seems that the video downlink from these drones has never been encrypted and it has been possible for those under surveillance to intercept and view the video feed.
This is the kind of news that makes great headlines. People read about it and slap their foreheads, proclaiming in a righteous voice, "What were they thinking? Head's should roll!" Stuff like that.
Here is an alternative viewpoint: this whole situation could just be a result of acceptable RISK MANAGEMENT practices.
I can imagine that there was an assessment of the risks (the PROBABILITY of a given bad thing happening) vs. the various costs (e.g., money, BAD PRESS) early in the development of the drone program. The estimated cost of the engineering was judged more than the estimated cost of adverse events.
Consider the following hypothetical scenario
Military Procurement to Corporate Executive
"We need to build an unmanned aerial vehicle and the price has to be under X dollars. We need it within Y months."
Corporate Executive to Engineer
"We only have X/2 dollars and Y/2 months to build this UAV thing."
Engineer to Corporate Executive
"There is no way in heck that we can encrypt the video downlink feed AND provide robust PKI. If we don't have good PKI then the system will be too hard to use. Doing it right will take twice as long and twice the money."
Corporate Executive to Military Procurement
"We can build what you requested. One caveat is the encrypted video stream - we can't meet your deadline and do that too, but we can get that in in the next version. Don't worry, though - the design includes protection/encryption of the control systems."
Military Procurement to Some General
"The control path is safe, so we don't have to worry about that. We need this in the field now because it will save lives. It is possible that the bad guys *might* be able to figure out where we are looking but it's unlikely that they can do so in time to run away."
Some General
"Fighting for more money will take time and is a pain in the keister. Make it happen."
This makes sense, right? The choices all seem reasonable. But here is the part that too often gets left out of the equation. :} These events take place months or years later.
Corporate Executive to Engineer
"Ok, it is time to build the next version of the UAV thing. We negotiated an extra Q/2 dollars to add in the encrypted video feed."
Engineer to Corporate Executive
"Well, that isn't enough. It would have cost Q/2 dollars to put it in first version. Now we have to retrofit the transmitters, modify all the receivers, and retrain all the personnel to use the new system, including how to use PKI. That will cost at least 4*Q dollars."
Corporate Executive
"!@#$%^&* - we have to have these other features! We'll have to wait on the encrypted video feed."
Now, years and years later, the players are still making choices based on weighing RISK vs COST. However, perhaps that RISK MANAGEMENT equation needed to include the true cost of not doing it right the first time, as well as the mathematical certainty that, if you keep rolling the dice over and over again, you will eventually observe even the least likely result.
The Predator video encryption snafu is in the news because the dice finally came up snake eyes. Even the best risk management practices do not provide a guarantee against a bad outcome, so be ready.
Reader Comments