When Machines See: Navigating Privacy
summary
Robots with always-on sensors collect thermal images, gait and facial data and upload it over networks, creating concrete vulnerabilities in data in motion and unclear ownership of captured identifiers. Field failures—robo taxis failing, a stuck Whimo, a farm helper called Truckucky—illustrate real reliability gaps. A robotics privacy framework requires three years to develop and prescribes privacy-by-design steps because passive, unconsented encounters replace button-click consent and expose biometric and behavioral traces to new attack surfaces.
AI-written summary · about
chapters (8)
- 0:00Privacy and robotics intro
- 3:41From civil engineering to robotics
- 6:13First robot encounters
- 8:04Nightcope robot privacy
- 10:34Rethinking autonomous privacy
- 12:01Building trust in robotics
- 13:27Robotics data security
- 14:11Embodied AI future
related talks
41mBeing Good: An Introduction to Robo- and Machine Ethics
27mThinking Ahead About AI Security and Privacy
42mThe rise of (evil) robots - cyber meeting safety in autonomous systems
59mPentesting AI: How to Hunt a Robot
47mDrones and Autonomous Vehicles: Privacy & Security vs. Surveillance
01h 05mNavigating the Security and Privacy Landscape of Modern AI
15mNavigating the Privacy Landscape of AI on the Devices: Challenges and Best Practices
47mSecuring Sexuality: Rewiring the Most Intimate Connections