Threat Modeling Thursday: 2018

Since I wrote my book on the topic, people have been asking me “what’s new in threat modeling?” My Blackhat talk is my answer to that question, and it’s been taking up the time that I’d otherwise be devoting to the series.

As I’ve been practicing my talk*, I discovered that there’s more new than I thought, and I may not be able to fit in everything I want to talk about in 50 minutes. But it’s coming together nicely.


The current core outline is:

  • What are we working on
    • The fast moving world of cyber
    • The agile world
    • Models are scary
  • What can go wrong? Threats evolve!
    • STRIDE
    • Machine Learning
    • Conflict

And of course, because it’s 2018, there’s cat videos and emoji to augment logic. Yeah, that’s the word. Augment. 🤷‍♂️

Wednesday, August 8 at 2:40 PM.

* Oh, and note to anyone speaking anywhere, and especially large events like Blackhat — as the speaker resources say: practice, practice, practice.

Half the US population will live in 8 states

That’s the subject of a thought-provoking Washington Post article, “In about 20 years, half the population will live in eight states,” and 70% of Americans will live in 15 states. “Meaning 30 percent will choose 70 senators. And the 30% will be older, whiter, more rural, more male than the 70 percent.” Of course, as the census shows the population shifting, the makeup of the House will also change dramatically.

Maybe you think that’s good, maybe you think that’s bad. It certainly leads to interesting political times. Maybe even a bit of chaos, emerging.

Keeping the Internet Secure

Today, a global coalition led by civil society and technology experts sent a letter asking the government of Australia to abandon plans to introduce legislation that would undermine strong encryption. The letter calls on government officials to become proponents of digital security and work collaboratively to help law enforcement adapt to the digital era.

In July 2017, Prime Minister Malcolm Turnbull held a press conference to announce that the government was drafting legislation that would compel device manufacturers to assist law enforcement in accessing encrypted information. In May of this year, Minister for Law Enforcement and Cybersecurity Angus Taylor restated the government’s priority to introduce legislation and traveled to the United States to speak with companies based there.

Today’s letter signed by 76 organizations, companies, and individuals, asks leaders in the government “not to pursue legislation that would undermine tools, policies, and technologies critical to protecting individual rights, safeguarding the economy, and providing security both in Australia and around the world.” (Read the full announcement here)

I’m pleased to have joined in this effort by Accessnow, and you can sign, too, at https://secureaustralia.org.au. Especially if you are Australian, I encourage you to do so.

Threat Modeling Thursday: 2018

So this week’s threat model Thursday is simply two requests:

  1. What would you like to see in the series?
  2. What would you like me to cover in my Blackhat talk, “Threat Modeling in 2018?”

“Attacks always get better, and that means your threat modeling needs to evolve. This talk looks at what’s new and important in threat modeling, organizes it into a simple conceptual framework, and makes it actionable. This includes new properties of systems being attacked, new attack techniques (like biometrics confused by LEDs) and a growing importance of threats to and/or through social media platforms and features. Take home ways to ensure your security engineering and threat modeling practices are up-to-date.”

Automotive Privacy

[Update: clarified a sentence about whose privacy is touched, and where.]

I had missed the story “Big Brother on wheels: Why your car company may know more about you than your spouse.” There are surprising details, including that you might be able to shut it off, and the phrase “If a customer declines, we do not collect any data from the vehicle.” I do wonder how a customer can decline — does it involve not buying a GM car?

When we did a privacy threat model at the Seattle Privacy Coalition, we found these issues. We also were surprised that the defense, taking a car driven by someone else (a taxi, or a Lyft/Uber) makes such a big difference, leaving the owner of the car associated with the trip via license plate, toll beacons, tire pressure monitors, traffic sensors, maps, and other technologies with tracking implications. And the passenger is associated if payment is by card, or the ride is booked via an app. splits/confuses the difference. It may also be that driving for Lyft/Uber acts as a defense, by classifying a car as a carshare, but it seems pretty easy to see through that to where the car is parked (especially overnight) and to repeated trips to dis-ambiguate between paid and personal rides.