Intel Hardware Security Academic Award
Rewarding Outstanding Innovation
The Intel Hardware Security Academic Award recognizes advancements in solutions, tools, and methodologies, which enhance the industry’s ability to deliver more secure and trustworthy foundational technologies. Launched in 2021, the program awards published papers that contain outstanding novel research with a meaningful impact on the hardware security ecosystem and industry, including technologies created by Intel.
The program also awards a Test of Time honor for research published more than 10 years ago that has demonstrated a significant and lasting impact in the security field.
Research Focus Areas
While we welcome all recently published papers demonstrating innovative research that advances product security in architecture, design, development, and validation, we highly encourage research in the following focus areas:
- Use of AI to accelerate or enhance security and security developed to protect AI technologies from a hardware standpoint.
- Use of analytics and machine learning to improve product security capabilities and robustness.
- Groundbreaking advancements in foundational security capabilities, including next-generation cryptographic techniques, safety-critical systems verification, and resilience against adversarial behavior.
- Innovations in scalable, automated tools and methodologies for hardware design and verification that are effective in addressing common security weaknesses, significantly improving product quality and assurance efficiency.
- Emerging usages, threat analysis, systemic mitigations, and security enhancements that strengthen cloud-to-edge computing, accelerant, and communication solutions.
- Architectural, micro-architecture, and circuit innovations that improve resiliency and reliability of silicon and electronics against transient faults.
- Innovations in Confidential Manufacturing methodologies, tools, and capabilities to support Intel’s IDM 2.0 vision, offering assurance, transparency, and a trusted supply chain to the ecosystem.
2022 Award Winners
First Place: “A Formal Approach to Confidentiality Verification in SoCs at the Register Transfer Level”
In this paper, researchers demonstrate how Unique Program Executing (UPEC) methodology can be used to reason about confidentiality properties of an SoC. UPEC methodology employs an efficient, induction-based formulation for information flow tracking. While the original UPEC methodology was formulated for micro-architectural side-channel detection for CPUs, this work is demonstrating how to generalize and scale that methodology for confidentiality properties for SoCs. Their formulation works directly on RTL and has been integrated in one commercial tool backend, thus yielding a first-of-its-kind, practically viable Pre-Si security verification technique.
Second Place: “Nyx: Greybox Hypervisor Fuzzing using Fast Snapshots and Affine Types”
This research demonstrates how modern hardware features (Intel® Virtualization Technology, extended page tables [EPT], Intel® Processor Trace, and page-modification logging [PML]) can be used to build effective and innovative security validation tools. The research has greatly improved the ability to test critical system software, ranging from embedded x86 firmware, to drivers, hypervisors, and future confidential compute stacks. In fact, within Intel’s Security Center of Excellence, researchers have already begun to leverage and evolve the technology, and the results have contributed to an even stronger software development lifecycle.
Test of Time Winner: “AEGIS: Architecture for Tamper-Evident and Tamper-Resistant Processing”
Published in 2003, this research describes a single-chip secure processor including a configuration where the underlying system software is untrusted. The proposed architecture incorporated several novel ideas at that time, such as cryptographic measurement and attestation, memory integrity verification, and memory encryption. This work helped inspire the broader domain of Trustworthy Computing and the practical realization of the novel features can be found in a wide range of Trusted Execution Environments (TEEs) deployed across the industry today.