对于抗量子零信任的人工智能安全性的分类框架
I Cherkaoui1, C Clarke2, J Horgan2
1Walton Institute, South East Technological University, Waterford, Ireland. Ilias.Cherkaoui@WaltonInstitute.ie.
Scientific reports
|February 3, 2026
概括
本研究介绍了一种结合后量子加密 (PQC) 和零信任架构 (ZTA) 的新框架,用于保护人工智能模型免受先进威胁. 该系统确保了强大的AI安全性,并提供了高效,可验证的保护.
科学领域:
- 计算机科学 计算机科学
- 网络安全 网络安全
- 人工智能的人工智能
背景情况:
- 快速部署人工智能需要先进的安全措施来应对不断发展的对抗性威胁.
- 现有的安全架构很难为人工智能模型提供强大的保护.
- 对于抗量子安全解决方案的需求至关重要.
研究的目的:
- 提出一个新的框架,将后量子加密 (PQC) 与零信任架构 (ZTA) 集成在一起,以确保AI模型访问.
- 为了在严格的安全证明中,在类别理论中正式奠定框架.
- 在实际实施中证明框架的有效性和效率.
主要方法:
- 将PQC原体与ZTA原则集成在一起.
- 正式建模密码工作流程作为形态和信任政策作为函数使用类别理论.
- 在ESP32平台上实施,以验证加密-敏捷过渡.
- 评估内存效率和访问控制性能.
主要成果:
- 该框架为基于格子的PQC提供细粒度,适应性信任和微细分类.
- 在加密-敏捷过渡中展示了可量化的改进.
- 实现了显著的内存效率:代理使用91.86%和经纪人使用97.88%的自由堆.
- 系统拒绝100%的未经授权的访问尝试与亚毫秒的延迟.
结论:
- 分类理论为人工智能安全提供了严格的数学基础.
- 拟议的PQC-ZTA框架加强了对抗AI威胁的保护.
- 实施验证了系统的效率,安全性和实际可用性.
相关概念视频
Quantum Numbers
50.5K
It is said that the energy of an electron in an atom is quantized; that is, it can be equal only to certain specific values and can jump from one energy level to another but not transition smoothly or stay between these levels.
50.5K
The Quantum-Mechanical Model of an Atom
57.7K
Shortly after de Broglie published his ideas that the electron in a hydrogen atom could be better thought of as being a circular standing wave instead of a particle moving in quantized circular orbits, Erwin Schrödinger extended de Broglie’s work by deriving what is now known as the Schrödinger equation. When Schrödinger applied his equation to hydrogen-like atoms, he was able to reproduce Bohr’s expression for the energy and, thus, the Rydberg formula governing hydrogen spectra.
57.7K
How Data are Classified: Categorical Data
44.7K
A variable, usually notated by capital letters such as X and Y, is a characteristic or measurement that can be determined for each member of a population. Data are the actual values of variables. They may be numbers, or they may be words. Datum is a single value.
Data are classified based on whether they are measurable or not. Categorical data cannot be measured; instead, it can be divided into categories. For example, if Y denotes a person's party affiliation, some examples of Y include...
Data are classified based on whether they are measurable or not. Categorical data cannot be measured; instead, it can be divided into categories. For example, if Y denotes a person's party affiliation, some examples of Y include...
44.7K
Resistivity
4.6K
When a voltage is applied to a conductor, an electrical field is generated, and charges in the conductor feel the force due to the electrical field. The current density that results depends on the electrical field and the properties of the material. In some materials, including metals at a given temperature, the current density is approximately proportional to the electrical field. In these cases, the current density can be modeled as:
4.6K
Resistance
6.0K
When a current moves through any conductor, the conductor causes some level of difficulty for the current to flow. The measure of that difficulty is known as the resistance of the material and is represented by R. Every material has its own resistance. In the case of conductors, heat is emitted whenever a current passes through them. Resistance depends on the resistivity of the material. Resistivity is a characteristic of the material used to fabricate electrical components, whereas the...
6.0K
Equivalent Resistance
985
In circuit analysis, situations often arise where resistors are neither in series nor parallel configurations. To tackle such scenarios, three-terminal equivalent networks like the wye (Y) (Figure 1 (a)) or tee (T) and delta (Δ) (Figure 1 (b)) or pi (π) networks come into play. These networks offer versatile solutions and are frequently encountered in various applications, including three-phase electrical systems, electrical filters, and matching networks.
985


