Cybersecurity Blog of Fraunhofer AISEC: OpenTitan Silicon Evaluation

Laboratory Security Evaluation of the OpenTitan Silicon Root of Trust

Together with Google and lowRISC®, Fraunhofer AISEC is investigating the hardware security of the OpenTitan chip (OpenTitan Silicon Root of Trust) and identifying key security vulnerabilities and hardening measures.

The OpenTitan Project

OpenTitan® is an open-source silicon root-of-trust (RoT) project that provides secure boot, key management, and essential cryptographic services. Developed by the OpenTitan partners and stewarded by lowRISC®, it is designed for use in high-assurance environments. The first OpenTitan-based silicon, manufactured by Nuvoton, implements the Earl Grey top level and is deployed in Chromebooks and soon in Google data centers to provide platform security. To test against the security requirements for these deployments, Fraunhofer AISEC conducted an in-depth laboratory security evaluation of OpenTitan.

Security Evaluation

The work was conducted in the Fraunhofer AISEC’s Common Criteria EAL 7-certified laboratory site, which is the highest evaluation assurance level (EAL) defined in the Common Criteria (CC) standard and meets the most stringent security requirements. This ensures that the evaluation follows processes and procedures that meet the highest standards required for products aiming for formal security certification. 

Fraunhofer AISEC’s Hardware Security Laboratory performs security analysis based on the latest state-of-the-art analysis methodologies with a focus on:

  • High-precision side-channel analysis,
  • Physical and semi-invasive attacks, including high-precision laser fault injection,
  • Optical emission analysis, and
  • Design and assessment of countermeasures against physical attacks.

In the laboratory, the Fraunhofer AISEC team is equipped with multiple setups for side-channel analysis, laser fault injection and electromagnetic fault injection.

These analysis techniques were applied to the OpenTitan engineering and production silicon to assess the robustness of its cryptographic engines, core components, and security-critical control logic against advanced physical attacks.

The evaluation was structured into several sub-stages, each focusing on specific OpenTitan components – such as the OpenTitan Big Number (OTBN) accelerator, memory, the Ibex core, and the cryptographic accelerators – and on their security properties. Across all sub-stages, the evaluation assumed very strong attacker capabilities with physical access to the device, including advanced side-channel and fault injection techniques. For each component, the evaluation focused on properties such as resistance against side-channel key extraction and fault-induced (control-flow) manipulation. This was complemented by optical analysis tools such as photon emission, which allowed the localization of the security critical components on the chip for the analysis.

The chip’s preparation to expose the silicon die for analysis through chemical etching and high-precision milling was also carried out in-house at the certified Fraunhofer lab facilities.

During the testing process, Fraunhofer AISEC worked with lowRISC, who provided silicon security expertise and customized pentesting firmware.

Open-Source Test and Evaluation Framework

Cybersecurity Blog of Fraunhofer AISEC: LFI set-up of the OpenTitan sample.
Figure 1: LFI set-up of the OpenTitan sample.

OpenTitan is not only open-source hardware; it also comes with an open-source test catalog for side-channel analysis and fault injection enabling third parties to:

  • Reproduce published measurement and analysis campaigns,
  • Extend existing tests or develop new ones, and
  • Adapt the infrastructure to their own targets and research needs.

The framework was developed and enhanced with the partners during the course of the collaboration.

Because the framework is developed in the open, the community can continuously refine the tools, share new analysis scripts, and enhance analysis methods over time. This toolkit formed the basis for the side-channel and fault-injection experiments in the campaign.

Impact on OpenTitan and Its Deployment

The collaboration delivered the following concrete outcomes:

  • Independent assessment of core security properties
    An independent security lab evaluated key OpenTitan design properties. Because OpenTitan is open-source, third parties review the design and reproduce these results, increasing confidence in using OpenTitan as the foundation of platform security.
  • Improved tooling and processes for ongoing assurance
    The open-source evaluation framework, together with improved debug and test hooks in OpenTitan, reduces the effort for future thorough audits, certifications, internal testing, and community review. The collaboration led to improvements in this framework.
  • Use case agnostic platform security
    The assessment shows that OpenTitan’s robust security model is applicable to a wide range of use cases. Its open-source design enables manufacturers to adapt and extend the architecture to meet their specific requirements.
  • Enhance research collaborations
    This work opens new opportunities for further research partnerships and collaborative innovation, including joint exploration of emerging technologies such as post-quantum cryptography.

This work demonstrates that the open-source root of trust – OpenTitan Earl Grey production silicon – integrated into servers and Chromebooks has been scrutinized using state-of-the-art side-channel and fault-injection techniques.

The evaluation was carried out in an environment designed for the highest levels of assurance. It shows that open hardware like OpenTitan can combine transparency, independent high-assurance evaluation, and practical deployability in real-world products.

We look forward to engaging with the OpenTitan Project more deeply in the future. Through the contributions from this joint collaboration we would like to call upon other OpenTitan or OpenTitan-based silicon manufacturers to use and continue to build on the contributions made to the open-source hardware testing infrastructure.

Cybersecurity Blog of Fraunhofer AISEC: LFI set-up of the OpenTitan sample.
Figure 2: LFI set-up of the OpenTitan sample.
Authors
Nisha_Jacob_Kabacki_Fraunhofer_AISEC_Hardware_Security_rund
Nisha Jacob Kabakci

Dr.-Ing. Nisha Jacob Kabakci studied electrical engineering in Bangalore and embedded systems at Università della Svizzera italiana in Lugano, complemented by a practical year in cryptography at Nanyang Technological University in Singapore. Since joining Fraunhofer AISEC, she has conducted research on the security of embedded systems with a focus on FPGAs, within which she completed her Ph.D. at the Technical University of Munich in 2020. She is currently Head of Department »Hardware Security – Physical Analysis and Countermeasures« at Fraunhofer AISEC.

Autor Cybersecurity Blog Fraunhofer AISEC
Marc Schink

Marc Schink carries out research in the field of »Hardware Security« at Fraunhofer AISEC. In his private life as well as at the institute, he strives to detect vulnerabilities in hardware and software. He has conducted several vulnerability disclosure processes with renowned and international manufacturers.

Most Popular

Never Miss a Post:

 
Bitte füllen Sie das Pflichtfeld aus.
Bitte füllen Sie das Pflichtfeld aus.
Bitte füllen Sie das Pflichtfeld aus.

* Mandatory

* Mandatory

By filling out the form you accept our privacy policy.

Leave a Reply

Your email address will not be published. Required fields are marked *

Other Articles

quantum computer with neural network visualization

Quantum Neural Network Security: Kill-Chain Attacks on QNNs

Quantum machine learning is moving from theory to real quantum hardware. This creates new security risks. In a recent study, researchers from Fraunhofer AISEC and partners demonstrated a multi-stage attack on a quantum neural network running on trapped-ion hardware. The attack combines side-channel analysis, adversarial examples and crosstalk. For providers, product teams and decision-makers, the message is clear: quantum AI needs security by design before it becomes a critical infrastructure.

Read More »

Fault Attacks on ECC Signature Verification

Digital signatures used in embedded systems are often based on elliptic curve cryptography (ECC) thanks to its performance and low memory profile. In secure boot processes they provide the cryptographic foundation for guaranteeing the authenticity of a firmware image. At the same time, these resource-constraints and the physical exposure of such devices makes them prime targets for fault attacks. Prior work studied faults on signature generation in depth, yet nobody had systematically asked how vulnerable signature verification is to fault attacks combined with carefully crafted signature inputs. That is exactly the gap we set out to close.

Read More »
Visual of the Blog Post: How Secure Is Europe's Rail System? A Systematic Cybersecurity Risk Analysis of the ERTMS

How Secure Is Europe’s Rail System? A Systematic Cybersecurity Risk Analysis of the European Rail Traffic Management System (ERTMS)

The European Rail Traffic Management System (ERTMS) is designed to make Europe’s railways interoperable and safe. The system, which has been mandatory since 2002, combines standardized signaling, radio communication, and train control – but how well is it protected against cyberattacks? Using MoRA, our modular risk assessment approach, we systematically model the ERTMS and evaluate cybersecurity risks. We compare current and future configurations to identify the assets with the largest attack surface. Our overview of the entire system – from GSM-R, balises, FRMCS, ETCS levels, and ATO to key management – provides a concise preview of risk profiles and attack tree scenarios, and highlights practical countermeasures that significantly improve cybersecurity.

Read More »
Viisual for blog post: Codyze: Automated Analysis of Cybersecurity Requirements in Software

Codyze: Automated Analysis of Cybersecurity Requirements in Software

Manually verifying compliance with requirements such as the Cyber Resilience Act is not scalable. Our code analysis tool, Codyze, translates product-centric regulatory requirements into verifiable rules and automatically assesses whether the product’s source code fulfills them – across languages and microservices. Codyze makes all analysis results transparent to developers, security teams, and auditors.

Read More »
WordPress Cookie Plugin by Real Cookie Banner