Authenticity is hard to find these days.

Disinformation has been a problem since the time of Ramses II and Julius Caesar (they both created false representations of their military achievements in monuments and the written word), and the rise of social media and the ease of AI-generated content has only exacerbated the problem. It is easy to create and spread an inauthentic account, recording, or video.

Combatting this rise in falsehood requires transparency and cooperation between publishers, creators, and consumers. The Coalition for Content Provenance and Authenticity ([C2PA](https://c2pa.org/)) was formed to address these challenges by developing an open standard that enables all parties to verify the origins and history of digital content.

C2PA is a useful tool in combatting disinformation, with sponsorship from over 6000 different organizations, and the steering committee currently includes Adobe, Amazon, BBC, Google, Intel, Meta, Microsoft, OpenAI, Publicis Groupe, Sony, and Truepic. It's viability is not in doubt.

This session will explore the current state of C2PA, examining (and demoing, where possible) open-source tooling available for use today. We'll also examine the discussion of privacy issues potentially presented by a standard that supports provenance—presenting both sides of the ongoing discussion within the community. 

Additionally, we'll also describe the role that C2PA might play in reducing the attack surface in AI and ML, exploring how Content Credentials (the artifact produced by the C2PA specification) can assist in mitigating data poisoning, software poisoning, or model poisoning attacks.