SPDM 1.3 and Beyond

Data Security

There has been much discussion about quantum-resilient cryptography (QRC), with a heavy focus on algorithms. With the NIST publication of the FIPS 203, FIPS 204, and FIPS 205 standards as well as adoption efforts underway in ISO/IEC, the focus is shifting to integration and use of QRC. Until recently, the Commercial National Security Algorithm Suite (CNSA) 2.0 Timeline was one of the few published timetables for transitioning to QRC. With the signing of Executive Order 14409, titled "Securing the Nation Against Advanced Cryptographic Attacks, the White House is accelerating the U.S. government’s transition to Post-Quantum Cryptography (PQC).

This session goes beyond the typical QRC algorithm summary and explores the issues and expectations that developers need to consider when preparing products and solutions to fully operate in PQC-ready environments.

Eric Hibbard

Director, Product Planning – Security / Chair, Security Technical Work Group

Samsung Semiconductor / SNIA

Data Security

Autonomous AI agents are rapidly evolving into primary consumers of enterprise data, operating across distributed computing environments where they plan tasks, invoke tools, and coordinate across clustered runtimes, network fabrics, and storage back ends at machine speed. As agents increasingly act as users of infrastructure rather than passive workloads, they introduce a new class of security challenges that remain poorly understood and insufficiently addressed. In this work, we analyze the emerging security implications of agent interoperability protocols, including Model Context Protocol (MCP), Agent-to-Agent (A2A), and agent commerce interfaces. We show that the combination of natural-language-driven control and machine-speed execution creates novel control-plane risks that existing runtime and infrastructure defenses are not designed to mitigate. 

We identify a set of critical vulnerability classes spanning both control and data planes, including protocol-level weaknesses in MCP and A2A, cross-protocol trust boundary failures, identity ambiguity at the data layer, privilege overreach at scale, audit and lineage gaps, and control-plane blindness under high-speed autonomous operation. Some of these risks are inherited from traditional distributed systems, while others are fundamentally amplified, or newly introduced, by agentic behavior. As agentic workloads become dominant consumers of enterprise storage and data, their security can no longer be addressed solely within AI model or application layers. Instead, it demands coordinated attention from storage and networking architects who define how data is accessed, governed, and secured at the infrastructure level. 

This session frames these challenges as an open problem space for SNIA developers and infrastructure architects, highlighting the need for new abstractions, controls, and standardization efforts at the storage and fabric layers.  Join us for an interactive discussion on these emerging threats with agentic data access and the requirements to protect enterprise systems as agents become first-class actors in distributed environments.

Sagar Kewalramani

Technical Staff, Engineering Technologist

Dell

Saoni Mukherjee

Senior Software Engineer

Dell

Data Security

Current data security measures are demonstrably not working in practice. Data breaches continue to occur because real systems are routinely deployed with misconfigurations that allow unauthorized access to data. Such errors are common, easy to make, and often high-impact.

There have been many attempts to solve this with new policy languages and enforcement engines that aim to decouple policy from code and enable uniform enforcement. In principle, centralization should allow consistent security decisions across services. In practice, however, writing even “simple” policies that correctly capture intent remains challenging: small mistakes can scale to organization-wide exposure.

We argue that this difficulty is structural. Modern policy languages (i) are too complex to reason about locally, because they admit highly expressive logical conditions whose global effects are hard to anticipate, and (ii) lack critical features that enable robust, collaborative policy development.

Without simple policy-language semantics and an explicit compositional structure, collaborative policy authoring inevitably introduces misconfigurations. In any organization of meaningful size, no single author controls the entire policy corpus; teams, departments, and data owners must each contribute rules that reflect their domain-specific requirements. Effective policy management demands effective collaboration, but current centralized approaches are nearly incompatible with it. Policies with high expressivity and unconstrained interactions have hard-to-anticipate consequences because every author must reason about every other author’s rules before deploying their own. This obligation scales poorly and is routinely violated.

At its core, the challenge is usability. Usability is often narrowly associated with interface design, but we adopt a broader view: usability is the fundamental property that enables users to operate a system correctly. A security mechanism that is difficult to use will be used incorrectly, and incorrect use is insecurity. In the context of access-control policy, correct operation means that individual authors can faithfully translate their own intent into function, and multiple authors can compose their efforts into a coherent and correct whole.

In this presentation, we will describe our open-source research project that aims to solve this problem. We will show how we decompose policy usability into two complementary dimensions: local and global. Local usability concerns the individual author’s experience: how easy is it to write, read, and understand my own policies? Local usability is addressed primarily through simpler semantics: a language whose constructs have clear, predictable meanings eliminates common single-author errors. 

Global usability concerns the collaborative setting: what must I know of other authors’ policies before I can safely deploy mine? In current systems, the answer is effectively all of it because of cascading interactions. Global usability is addressed primarily through separation of concerns: a compositional structure that bounds the effects of each author’s policies to a well-defined scope, so that the private actions of one author cannot break the work of another.

Our focus is data access management, where the stakes of misconfiguration are especially high, and the need for multi-stakeholder collaboration is acute. We propose a policy language that is simultaneously radically simpler in form and designed from the ground up to support collaborative authoring.

Ted Dunning

Fellow

Hewlett Packard Enterprise

Jonah Rosenblum

PhD Candidate / Research Associate

University of Michigan / Hewlett Packard Enterprise

Data Security

The emergence of quantum computers is presenting new security challenges for storage and memory subsystems.  Timelines are established for increasing the security levels, protocols, and inventory traceability of components and assemblies that are exposed to new sophisticated attacks.   These changes not only affect the designs of new chips, but also include validating the supply chain dynamically.   New encryption algorithms are coupled with software and hardware bills of materials to establish a security regime that deflects threats and keeps our computing environment safe.

Bill Gervasi

Principal Memory Solutions Architect / Chair, Emerging Memories

Monolithic Power Systems / JEDEC