The rise of (evil) robots - cyber meeting safety in autonomous systems
summary
Autonomous Mobile Robots already move payloads from ~100 kg up to 2 tonnes and attackers shift from economics to physical harm by tampering control and calibration to cause DoS, faulty parts, or collisions. Real-world vectors include firmware rollback, unsigned updates, embedded credentials, Wi‑Fi/5G/GPS/Bluetooth spoofing, and sensor attacks on camera/LiDAR/GPS; examples cited include Stuxnet-style OT sabotage and hospital robot hijacks. Defenses are anti-rollback and secure boot, signed calibration data, network segregation, IDS/monitoring, SBOM/SCA, OS hardening, least-privilege accounts and disabled unused interfaces. Safety requires security and shared responsibility across vendor, integrator and operator.
AI-written summary · about
chapters (16)
- 0:00Introduction and speaker background
- 2:33Industrial revolutions overview
- 9:02Autonomous mobile robots explained
- 12:58Real-world robot hacking examples
- 17:58Attack goals and risks in AMRs
- 19:45Firmware and software attack vectors
- 21:10Networking and communication risks
- 22:57Hardware and physical security threats
- 24:45Supply chain and software security
- 27:17Sensor vulnerabilities and defenses
- 29:27Identity and access management flaws
- 31:32Security vs safety in autonomous systems
- 35:52Shared responsibility for security
- 36:57AI's role in autonomous systems
- 39:06Secure by design principles
- 40:33Summary and future outlook
related talks
47mSecurity Architecture for Robotic and Autonomous Systems
41mPanel: Tackling Automotive Hardware Vulnerabilities
58mSecuring a World of Physically Capable Computers
21mIndustrial IoT security issues and countermeasures
19mSecuring the future of mobility: Automotive threat modeling
41mBeing Good: An Introduction to Robo- and Machine Ethics
31mHacking Everything – The Dark Side of the Internet of Things
20mAI Red Teaming and Automation: Exploring Societal Risks in...