FMEA-AI: AI fairness impact assessment using failure mode and effects analysis

Li, Jamy; Chignell, Mark

doi:10.1007/s43681-022-00145-9

Cited by 14 publications

(5 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AI failure modes typically lack consideration of industrial context and specific trustworthy requirements. For example [31], define failure modes in IT safety, while FMEA has been applied to TAI, albeit with limited success in addressing only one requirement such as fairness [36].…”

Section: Methodsmentioning

confidence: 99%

TAI-PRM: trustworthy AI—project risk management framework towards Industry 5.0

Vyhmeister,

Castane

2024

AI Ethics

View full text Add to dashboard Cite

Artificial Intelligence (AI) is increasingly being used in manufacturing to automate tasks and process data, leading to what has been termed Industry. 4.0. However, as we move towards Industry 5.0, there is a need to incorporate societal and human-centric dimensions into the development and deployment of AI software artefacts. This requires blending ethical considerations with existing practices and standards. To address this need, the TAI-PRM framework has been developed. It builds upon established methods, such as Failure Mode and Effect Analysis (FMEA) and the Industrial ISO 31000, to manage risks associated with AI artefacts in the manufacturing sector. The framework identifies ethical considerations as hazards that can impact system processes and sustainability and provides tools and metrics to manage these risks. To validate the framework, it was applied in an EU project for Digital Twins on AI for manufacturing. The results showed that TAI-PRM can effectively identify and track different failure modes associated with AI artefacts and help users to manage ethical risks associated with their deployment. By incorporating ethical considerations into risk management processes, the framework enables the developing and deploying trustworthy AI in the manufacturing sector.

show abstract

Section: Methodsmentioning

confidence: 99%

TAI-PRM: trustworthy AI—project risk management framework towards Industry 5.0

Vyhmeister,

Castane

2024

AI Ethics

View full text Add to dashboard Cite

show abstract

“…Yajima et al [34] showcased their work in progress on assessing machine learning security risks. Failure mode and effect analysis (FMEA) has been adopted/extended for assessing RAI risks in [35]- [37].…”

Section: Related Workmentioning

confidence: 99%

Towards Concrete and Connected AI Risk Assessment (C$^2$AIRA): A Systematic Mapping Study

Xia¹,

Lu²,

Perera³

et al. 2023

Preprint

View full text Add to dashboard Cite

The rapid development of artificial intelligence (AI) has led to increasing concerns about the capability of AI systems to make decisions and behave responsibly. Responsible AI (RAI) refers to the development and use of AI systems that benefit humans, society, and the environment while minimising the risk of negative consequences. To ensure responsible AI, the risks associated with AI systems' development and use must be identified, assessed and mitigated. Various AI risk assessment frameworks have been released recently by governments, organisations, and companies. However, it can be challenging for AI stakeholders to have a clear picture of the available frameworks and determine the most suitable ones for a specific context. Additionally, there is a need to identify areas that require further research or development of new frameworks. To fill the gap, we present a survey of 16 existing RAI risk assessment frameworks from the industry, governments, and non-government organizations (NGOs). We identify key characteristics of each framework and analyse them in terms of RAI principles, stakeholders, system lifecycle stages, geographical locations, targeted domains, and assessment methods. Our study provides a comprehensive analysis of the current state of the frameworks and highlights areas of convergence and divergence among them. We also identify the deficiencies in existing frameworks and outlines the essential characteristics a concrete framework should possess. Our findings and insights can help relevant stakeholders choose suitable RAI risk assessment frameworks and guide the design of future frameworks towards concreteness.

show abstract

“…FMEA first emerged in the military domain and then spread to the aerospace industry and to other manufacturing domain, with various applications in the nuclear electronics, and automotive fields as well. Recently, researchers have explored how FMEA or other safety engineering tools can be used to assess the design of AI-ML systems [41,42]. Applying FMEA to an AI-ML asset includes the following activities: (i) assigning functions to the asset, (ii) creating structure, function, networks diagrams for the asset, (iii) define defects that can cause the asset's function/function network to fail, (iv) perform threat modeling actions.…”

Section: Guide To Fmea Application In the Ai-ml Life-cyclementioning

confidence: 99%

Modeling Threats to AI-ML Systems Using STRIDE

Mauri

Damiani

2022

Sensors

View full text Add to dashboard Cite

The application of emerging technologies, such as Artificial Intelligence (AI), entails risks that need to be addressed to ensure secure and trustworthy socio-technical infrastructures. Machine Learning (ML), the most developed subfield of AI, allows for improved decision-making processes. However, ML models exhibit specific vulnerabilities that conventional IT systems are not subject to. As systems incorporating ML components become increasingly pervasive, the need to provide security practitioners with threat modeling tailored to the specific AI-ML pipeline is of paramount importance. Currently, there exist no well-established approach accounting for the entire ML life-cycle in the identification and analysis of threats targeting ML techniques. In this paper, we propose an asset-centered methodology—STRIDE-AI—for assessing the security of AI-ML-based systems. We discuss how to apply the FMEA process to identify how assets generated and used at different stages of the ML life-cycle may fail. By adapting Microsoft’s STRIDE approach to the AI-ML domain, we map potential ML failure modes to threats and security properties these threats may endanger. The proposed methodology can assist ML practitioners in choosing the most effective security controls to protect ML assets. We illustrate STRIDE-AI with the help of a real-world use case selected from the TOREADOR H2020 project.

show abstract

FMEA-AI: AI fairness impact assessment using failure mode and effects analysis

Cited by 14 publications

References 47 publications

TAI-PRM: trustworthy AI—project risk management framework towards Industry 5.0

TAI-PRM: trustworthy AI—project risk management framework towards Industry 5.0

Towards Concrete and Connected AI Risk Assessment (C$^2$AIRA): A Systematic Mapping Study

Modeling Threats to AI-ML Systems Using STRIDE

Contact Info

Product

Resources

About