en.Wedoany.com Reported - On June 3, Infineon Technologies announced the integration of the OPTIGA TPM SLB 9672 into the NVIDIA Jetson Thor platform, providing a hardware-based root of trust for robotics and autonomous systems. Designed for Physical AI devices, this solution focuses on key protection, system integrity verification, remote attestation, and post-quantum security upgrades.

This integration addresses the security foundation issue as robots transition from laboratories to large-scale deployment. Once industrial robots, logistics robots, medical assistance devices, and humanoid robots enter factories, warehouses, hospitals, and public spaces, their security risks are no longer limited to data breaches but also affect device operations, production continuity, personnel safety, and compliance responsibilities. As a computing platform for robotics and edge AI, Jetson Thor requires a verifiable security foundation for device startup, software updates, AI model key storage, communication encryption, and lifecycle management. By integrating the TPM security chip, which operates independently of the application processor, into this platform, Infineon enables developers to store encryption keys at the chip level and verify the authenticity, integrity, and untampered state of the software stack through measured boot and remote attestation.

This TPM solution supports hardware-based protection for AI model keys, encrypted communications, and signed over-the-air updates.

The uniqueness of Physical AI devices lies in their need to perceive real-world environments, execute physical actions, and continuously connect to the cloud, edge platforms, and enterprise management systems. As robots transition from single-unit pilots to fleet deployments, enterprises must not only focus on computing power, sensors, actuators, and power management but also ensure that each device maintains a consistent trusted identity, traceable software state, and auditable security update path over long-term operation. Infineon emphasizes that the OPTIGA TPM features a physically isolated design and holds FIPS and Common Criteria certifications, providing a security root independent of the main processor. Its firmware update mechanism introduces post-quantum security protection, with the future roadmap covering the ML-KEM and ML-DSA algorithms standardized by NIST in 2024. For robot manufacturers, whether such security capabilities are reserved during the early architecture phase will directly impact the ability of subsequent products to comply with the EU Cyber Resilience Act, the EU Artificial Intelligence Act, IEC 62443 industrial system security requirements, and compliance trends in industries such as healthcare and automotive.

Infineon also views the semiconductor needs for robotics within a more comprehensive system chain. Humanoid robots require multiple types of chips—including sensing, actuation, power management, connectivity, and security—to work in coordination. The company estimates that the semiconductor content per humanoid robot is approximately $500, with security components set to occupy a more critical position as regulatory requirements mature. As ecosystem players like NVIDIA and Infineon integrate around robotics platforms, robot security is shifting from software patches and external protections to the chip level, boot chain, device identity, and long-term update mechanisms. For Physical AI devices to enter demanding scenarios such as industrial, medical, and logistics applications, hardware-level security roots will become a foundational element that must be addressed before large-scale deployment.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com