A Taste Of Privacy Threat Modeling by Kim Wuyts
summary
Claims a practical four-step process (what, what can go wrong, fixes, review) and offers the Linden privacy threat modeling knowledge base (trees, cards). Cited findings: analysis of 1.4M Strava activities reveals home addresses for 85 percent of users; Wi‑Fi/Bluetooth voice links can be intercepted (Bluetooth eavesdrop 15 meters); a networked doll is banned in Germany for voice telemetry. Concrete mitigations include minimization, risk prioritization and on-device speech-to-text or scrambled analytics rather than raw uploads.
AI-written summary · about
chapters (17)
- 0:00The ice cream analogy
- 3:24Introduction to privacy
- 5:54Why privacy matters
- 8:35Data aggregation risks
- 11:37Privacy benefits and laws
- 14:15Defining privacy concepts
- 16:51Unlinkability and threats
- 19:44Privacy vs security
- 22:55Privacy by design
- 26:18Threat modeling framework
- 29:02Analyzing a smart toy
- 31:07Linden privacy framework
- 34:06Mitigation and solutions
- 36:58Successful threat modeling
- 39:59Future of threat modeling
- 43:39Final ice cream wisdom
- 45:05Question and answer session
related talks
44mA Taste of Privacy Threat Modeling
22mPrivacy Threat Modeling
48mStriding Your Way to LINDDUN: Threat Modeling for Privacy
43mThreat modeling
01h 00mThreat Modeling for Secure Software Design
17mPrivacy Design Flaws
26mThreat Modeling for IoT
41mPrivacy Threat Modeling Analysis of Cloud Services Against Privacy Regulations