everyone: older news
Impossible Problems in Computer Security
by Brent Kirkpatrick
(Date Published: 9/18/2018.)
DARPA oversold their Cyber Grand Challenge. Their web site says that the goal of their Cyber Grand Challenge was to create "automated defense systems capable of reasoning about flaws" or vulnerabilities in computer programs. They go on to describe this effort as a "breakthrough". Indeed, they are trying to solve an algorithmically intractable problem, that of detecting buffer over-run vulnerabilities. They oversold solutions to their challenge, because they failed to characterize the difficulties involved.
DARPA's imagined breakthrough will not happen. The math proves that their goal is impossible to achieve with an algorithm. The following problems are intractable for algorithms:
There are no silver bullets. The search for silver bullets is whimsical at best. I would encourage all of my colleagues to work on problems that are tractable. Computer security should be done by solving one small, tractable problem at a time.
Please contact us at Intrepid Net Computing if you need solutions to cyberattacks.
Trojan Hunter (TM). Digital forensics for Trojans at an accessible, fixed price. For any operating system.
Dustin Fraze. Cyber Grand Challenge (CGC). https://www.darpa.mil/program/cyber-grand-challenge. Accessed: Sept. 17, 2018.
Brent Kirkpatrick. Computer Security is Algorithmically Intractable. (whitepaper) 2018.
Brent Kirkpatrick. Computer Security is Algorithmically Intractable. Aug 16, 2018