Page 66 - AI Standards for Global Impact: From Governance to Action

AI Standards for Global Impact: From Governance to Action

Figure 34: AI and Multimedia Authenticity Standards Collaboration leadership team

AMAS work is centred around three pillars:

a) Technical Activities - This pillar is working on the mapping of the stan dardization landscape
in the areas of AI and multimedia authenticity, including the identification of gaps where
standards are still needed to support relevant policies across governments, industries, and
other stakeholders want to implement.
b) Policy – A forum for connecting with governments to discuss the alignment of policies and
standards and related lessons learned.
c) Communications - Promoting the work of AMAS and its value to various stakeholders (e.g,
governments, social media platform providers, media companies, consumers, etc.)

Two AMAS papers were presented at the summit:

i) Technical Paper on "Mapping the Standardization Landscape" developed by the Technical
Activities pillar
ii) Policy Paper on "Building Trust in Multimedia Authenticity through International Standards"
developed by the Policy pillar

10�2 Standards mapping landscape on AI and multimedia authenticity

The technical paper on AI and Multimedia Authenticity Standards provides a comprehensive
overview of the current landscape of standards and specifications related to digital media
authenticity (covering different media types from images, audio, video, data, and PDF, among
others). It categorizes these standards into five key clusters: content provenance, trust and
authenticity, asset identifiers, rights declarations, and watermarking. The report provides a short
description of each standard along with link for further details.

By mapping the contributions of various standards bodies and groups, the paper aims to identify
gaps and opportunities for further standardization. Over 35 technical standards from SDOs are
analysed in the standards mapping. This could serve as a valuable resource for stakeholders
seeking to navigate the complex ecosystem of standards at the intersection of AI and authenticity
in digital media and to implement best practices to safeguard the authenticity of digital assets
and the rights assigned to them. The findings underscore the critical role of robust specifications
and standards in fostering trust and accountability in the evolving digital landscape.

54