[Initiative]:Cloud Native AI Security Whitepaper - Model and Data Security

[dehatideep]

Name

Cloud Native AI Security Whitepaper - Model and Data Security

Short description

This whitepaper discusses model and data security in a CNAI environment

Responsible group

TOC

Does the initiative belong to a subproject?

Yes

Subproject name

AI

Primary contact

@deepcpat [Deep Patel, [email protected]]

Additional contacts

@joshhalley [Josh Halley] @raravena80 [Ricardo Aravena]

Initiative description

Original AI Security Whitepaper (https://github.com/cncf/toc/issues/1718) has been split into 3 parts as per suggestion of TOC and https://github.com/cncf/toc/issues/1718 is abandoned because it was a document having 50+ pages.

Three split docs issues are: https://github.com/cncf/toc/issues/1939 [Encryption Landscape] https://github.com/cncf/toc/issues/1940 [Model and Data Security] https://github.com/cncf/toc/issues/1942 [AI Agents]

This part covers Cloud Native AI Security and issues related to model and data security. It covers what could go wrong with model security and what defense could be deployed ti mitigate these. It also discusses data security at-rest, in-transit, and in-use. . Cloud Native AI Security Whitepaper - Model and Data security

Deliverable(s) or exit criteria

Document is ready and reviewed by TOC Subproject AI folks, given original AI security paper was ready and already submitted to TOC. As per advice of TOC original paper has been split into smaller docs having 10-12 pages. This part deals with model and data security. This is ready to be taken by TOC. Cloud Native AI Security Whitepaper - Model and Data security

Tracking document for meeting and progress

https://docs.google.com/document/d/1uggG2_hI7FgJMbsBEuOkbFla-E4pwTufYQbtGhSlWTM/edit?usp=sharing