AI的安全与隐私问题考量.pdf

《AI的安全与隐私问题考量.pdf》由会员分享，可在线阅读，更多相关《AI的安全与隐私问题考量.pdf（14页珍藏版）》请在三个皮匠报告上搜索。

1、1|2024 SNIA.All Rights Reserved.Security and Privacy Concerns for AI Eric Hibbard,CISSP,FIP,CISASamsung Semiconductor,Inc.2|2024 SNIA.All Rights Reserved.IntroductionArtificial intelligence(AI)systems are creating numerous opportunities and challenges for many facets of society.For security,AI is pr

2、oving to be a power tool for both adversaries and defenders.Privacy is similar,but the societal concerns are elevated to a point where laws and regulations are already being enacted.3|2024 SNIA.All Rights Reserved.AI Ethical and Societal Concerns Ethical and societal concerns are a factor when devel

3、oping and using AI systems and applications Taking context,scope and risks into consideration can mitigate undesirable ethical and societal outcomes and harms such as:financial harm psychological harm harm to physical health or safety intangible property(for example,IP theft,damage to a companys rep

4、utation)social or political systems(for example,election interference,loss of trust in authorities)civil liberties(for example,unjustified imprisonment or other punishment,censorship,privacy breaches)Source:ISO/IEC TR 24368:2022 Information technology Artificial Intelligence Overview of ethical and

5、societal concerns4|2024 SNIA.All Rights Reserved.Examples of potential harms related to AI systemsSource:NIST Artificial Intelligence Risk Management Framework(AI RMF 1.0)5|2024 SNIA.All Rights Reserved.OECD and Artificial IntelligenceThe OECD AI PrinciplesOECD Framework for the Classification of AI

6、 SystemsSource:OECD(2022)OECD Framework for the Classification of AI systems OECD Digital Economy PapersThe OECD has developed a framework for classifying AI lifecycle activities according to five key socio-technical dimensions,each with properties relevant for AI policy and governance,including ris