I am a security engineer for a local NC based insulation manufacturer. While doing my daily due diligence, I came across this VERY interesting security document. Even though Omnipod was not mentioned, I am calling them Monday to see what their path for securing against something like this.
Scary!
Tom
I saw that too. Basic things from the past apply here. We are a "needle in a haystack" and attacks like this are targeted. So the risk is so low it's not measurable. I'm not worried about it at all.
Are people sick or bored that they have to hack into a life saving device???? What the hell is wrong with people nowadays??? I can only hope those of us with children are teaching them values, ethics and a sense of work instead of tv and video games. This is awful!
Well we don't have to worry about this threat anymore:
And besides it is a very targeted attack.
" Baarnaby Jack, a hacker who was due to present his findings on the security vulnerabilities of implanted medical devices, has died.
The San Francisco medical examiner's office said Jack, 35, died in the city on Thursday – but did not provide details on the circumstances surrounding his death.
Jack had exposed a security flaw in insulin pumps that could be made to dispense a fatal dose by a hacker 300ft away, pushing some medical companies to review the security of these devices.
Pretty damn scary! I'd like to know what Insulet has to say.
OMG, he just died suddenly? Doesn't that seem strange??
Yeah, the timing is very strange. Especially since the cause of death is still unknown or hasn't been released yet if it is known.
It was discussed before. We don't know what form of encryption the Omnipod uses. It's easy to find the bandwidth it uses by reading the instructions (page 149 for the old pod, page 153 for the new) and this can be verified with a scanner.
We do know that the protocol to establish communication is extremely secure. IMO the old pod was extremely secure. Breaking it required a man-in-the-middle attack of some description and even then I can't see how to do it (outside "Mission Impossible" style approaches). With the new pod the pod and PDM have to be in contact; that only leaves the Mission Impossible approach (swap out the diabetic's supply of pods, or maybe the PDM, for special ones.)
So what this leaves is eavesdropping during setup to grab the security information before it starts being used. In principle this was possible with the old pods - they always transmitted with 16mW of power. The new pods transmit similar power (13mW) on a different band, however I believe the transmission power during setup, when security information would be transmitted, is far less (maybe 100 times less?)
My belief is that if the initial setup *is* secure then then it is *possible* to make the PDM-pod communication completely unbreakable for the pod life (3 days), simply by using a large enough code book. It is *possible* to make the communication effectively unbreakable by using a smaller code book and appropriate cryptography (e.g. just CBC on individual communications, so, trivially, the code book need only be as large as the *number* of successful communications.)
Of course, I don't know how Insulet does encode the communications, but I suspect a major justification for reducing the pod-PDM distance during setup was to address the issue of snooping during setup.
John Bowler
I'm glad I came back to read this. It's been freaking me out for days. I think as a New Yorker, I'm still a little traumatized by 9-11 and this did a number on me. Thanks for your thoughtful responses
Not for nothing, but this article is about 18 months old. I'm pretty sure if someone wanted to off me, they could find an easier method :)
I don’t think the article being a year and a half old makes it a non-issue. the fact that the guy died at 35 right before the demonstration and te fact that the pacemaker and pump companies haven’t responded to the claims in all that time is even stranger. but the facts laid out in this thread are what put my mind more at ease!
I saw this on DiabetesMine and thought it was applicable: http://www.diabetesmine.com/2013/08/diabetes-device-hacker-joins-forces-with-fda.html :) Interesting to see what will become of it all.
Maybe the FDA gunned him down for having an insulin pump without a prescription. :)
thank you! 
Wonder if he "hacked" himself? :) Sorry, just had to think it...
Pumps and monitors are hackable because there is no encryption. Data is transmitted plaintext because devices are small and lack processing power. However, there is a limit to what people can do because these are not IoT devices - at least, my devices DO NOT connect to the internet, so any hacker would have to be in a close geographic proximity. If you go to a large hacking conference, disconnect from your devices - go onto syringe insulin. Otherwise, you are probably fine. There ARE concerning security issues, but this is one of the lesser ones. BlackHat 2011 - Hacking Medical Devices for Fun and Insulin: Breaking the Human SCADA System - YouTube