Internship title: Security of embedded artificial intelligence system Thales
Meudon, {}

Job Description

Internship title: Security of embedded artificial intelligence system

Job ID: 18002668 Location: Meudon, Île-de-France Region, France

Full time, Internship / Placement

Trusted Labs is a global expert in security consulting and evaluation within the connected ecosystem and a world leader in security certification scheme definition. We have 17 year of expertise in securing and enhancing trust in the connect world and have a strong multi-sector expertise in many segments e.g. banking, automotive, government and energy. Moreover, Trusted Labs evaluation lab is accredited by the French Certification Body (Common Criteria and CSPN) as well as several proprietary schemes.

We are interested in candidates with a background in security and who are motivated on a variety of innovative security issues or topics.

The goal of this internship is to consider the security aspect of AI-based embedded systems: review the state of the art, identify potential risks and propose a framework to evaluate the security robustness of an AI-based embedded device.

Description

Artificial intelligence has evolved rapidly from an interesting research topic to an effective solution for a wide range of applications. Even though, generally speaking, AI services are deployed in the cloud which has significant advantages regarding computing capacity, embedded AI architecture (or the hybrid architecture) could be preferred for the following scenarios: real-time needs, connectivity, privacy and security concerns…

Embedded Artificial Intelligence gives smart devices an ability to be aware of its surroundings and make decisions on the received input. Due to the specific constraints of embedded devices, many actual works are focusing on the optimization of power, computing capacity and cost of AI-based embedded devices.

The goal of this internship is to consider another aspect of AI-based embedded systems: the security. It will start with a state of the art of security issues of AI-based embedded devices, followed by a risk analysis in order to identify potential risks. The final step will be to define a framework to evaluate the security robustness of an AI-based embedded device.