{"id":54749,"date":"2023-11-28T14:10:17","date_gmt":"2023-11-28T14:10:17","guid":{"rendered":"https:\/\/packetstormsecurity.com\/news\/view\/35238\/4-Key-Takeaways-From-New-Global-AI-Security-Guidelines.html"},"modified":"2023-11-28T14:10:17","modified_gmt":"2023-11-28T14:10:17","slug":"4-key-takeaways-from-new-global-ai-security-guidelines","status":"publish","type":"post","link":"https:\/\/www.threatshub.org\/blog\/4-key-takeaways-from-new-global-ai-security-guidelines\/","title":{"rendered":"4 Key Takeaways From New Global AI Security Guidelines"},"content":{"rendered":"
<\/div>\n

AI security guidelines developed by the United States’ Cybersecurity and Infrastructure Security Agency (CISA) and the United Kingdom\u2019s National Cyber Security Centre (NCSC) were published Monday<\/a> with endorsements from 16 other nations. The 20-page document was written in cooperation with experts from Google, Amazon, OpenAI, Microsoft and more, and is the first of its kind to receive global agreement, according to the NCSC<\/a>.<\/p>\n

\u201cWe know that AI is developing at a phenomenal pace and there is a need for concerted international action, across governments and industry, to keep up,\u201d said NCSC CEO Lindy Cameron, in a public statement. \u201cThese guidelines mark a significant step in shaping a truly global, common understanding of cyber risks and mitigation strategies around AI to ensure that security is not a postscript to development but a core requirement throughout.\u201d<\/p>\n

Here are four key takeaways from the publication:<\/p>\n

1. “Secure-by-design” and “secure-by-default” take priority<\/h2>\n

Emphasized throughout the document are the principles of \u201csecure-by-design\u201d<\/a> and \u201csecure-by-default\u201d \u2014 proactive approaches to protect AI products from attack. The authors urge AI developers to prioritize security alongside function and performance throughout their decision-making, such as when choosing a model architecture or training dataset. It is also recommended that products have the most secure options set by default, with the risks of alternative configurations clearly communicated to users. Ultimately, developers should assume accountability for downstream results and not rely on customers to take the reins on security, according to the guidelines.<\/p>\n

Key excerpt: \u201cUsers (whether \u2018end users,\u2019 or providers incorporating an external AI component) do not typically have sufficient visibility and\/or expertise to fully understand, evaluate or address risks associated with systems they are using. As such, in line with \u2018secure by design\u2019 principles, providers of AI components should take responsibility for the security outcomes of users further down the supply chain.\u201d<\/em><\/p>\n

2. Complex supply chains require greater diligence<\/h2>\n

AI tool developers frequently rely on third-party components like base models, training datasets and APIs when designing their own product. An extensive network of suppliers creates a greater attack surface where one \u201cweak link\u201d can negatively impact the product\u2019s security. The global AI guidelines recommend developers assess these risks when deciding whether to acquire components from third parties or produce them in-house. When working with third parties, developers should vet and monitor the security posture of suppliers, hold suppliers to the same security standards as one\u2019s own organization and implement scanning and isolation of imported third-party code, the guidelines state.  <\/p>\n

Key excerpt: \u201cYou are ready to failover to alternate solutions for mission-critical systems, if security criteria are not met. You use resources like the NCSC\u2019s Supply Chain Guidance<\/a> and frameworks such as Supply Chain Levels for Software Artifacts (SLSA) for tracking attestations of the supply chain and software development life cycles.\u201d<\/em><\/p>\n

3. AI faces unique risks<\/h2>\n

AI-specific threats<\/a> such as prompt injection attacks and data poisoning call for unique security considerations, some of which are highlighted by CISA and NCSC in their guidelines. A component of the \u201csecure-by-design\u201d approach includes integrating guardrails around model outputs to prevent leaking of sensitive data and restricting the actions of AI components used for tasks such file editing. Developers should incorporate AI-specific threat scenarios into testing and monitor user inputs for attempts to exploit the system.<\/p>\n

Key excerpt: \u201cThe term \u2018adversarial machine learning\u2019 (AML), is used to describe the exploitation of fundamental vulnerabilities in ML components, including hardware, software, workflows and supply chains. AML enables attackers to cause unintended behaviors in ML systems which can include:<\/em><\/p>\n

\n