An Emerging Threat to AI that Could Make Vehicles Crash and Facial Recognition Fail

[Brownie2019]

Content source

https://www.securityweek.com/oneflip-an-emerging-threat-to-ai-that-could-make-vehicles-crash-and-facial-recognition-fail/

Autonomous vehicles and many other automated systems are controlled by AI; but the AI could be controlled by malicious attackers taking over the AI’s weights.

Weights within AI’s deep neural networks represent the models’ learning and how it is used. A weight is usually defined in a 32-bit word, and there can be hundreds of billions of bits involved in this AI ‘reasoning’ process. It is a no-brainer that if an attacker controls the weights, the attacker controls the AI.

Full Story:

OneFlip: An Emerging Threat to AI that Could Make Vehicles Crash and Facial Recognition Fail

OneFlip is a Rowhammer-based attack that flips a single bit in neural network weights to stealthily backdoor AI systems.

www.securityweek.com

 

[Divergent]

Scareware tactics as per the norm. It's possibly but highly unlikely.

This conclusion is based on the two stringent requirements for the attack: (1) white-box access to the model and (2) co-location of the attacker's code on the same physical machine.