Protect AI models at the edge

Embedded devices increasingly rely on AI models to perform disparate tasks. Companies need to spend a lot of effort, time, and money to build AI models, such as deep neural networks, due to a long process that includes onerous design and development phases. AI models are however susceptible to theft, tampering, and unauthorized access due to the vulnerabilities of hardware and software environments. Concerning theft, when no proper countermeasures are in place, attackers could steal intellectual property that required enormous investments to be generated.

Protecting AI models on embedded devices while maintaining a seamless usage workflow is hence a major challenge. AI Bunker addresses this by enabling secure storage of AI models within a dedicated, strongly-isolated software partition on the embedded device, ensuring high levels of protection. At the same time, it transparently allows invoking the model’s inference using a standard workflow, making it appear as if the model resides in a regular environment.