C2PA standards body for AI: What marketers need to know

Posted On 08 Jun 2024

Comment: Off

US generative AI use has climbed from 7.8 million people in 2022 to 100.1 million people in 2024, per our forecast. Widely adopted AI use may mean increased productivity, but it also increases the risk for deepfakes—multimedia content that looks real but is AI generated. That’s why marketers need to pay attention to the Coalition for Content Provenance and Authenticity (C2PA).

The C2PA is a collection of representatives from companies including OpenAI, Google, Publicis, Microsoft, and Adobe with a goal to “make the world safe for generative AI,” according to Adam Buhler, steering committee member for the C2PA and executive vice president and head of creative technology at Publicis.

The group has created a standard that displays a piece of content’s origin. “It provides transparency into where that digital media comes from,” said Ray Lansigan, marketing and communications chair for the C2PA and executive vice president of corporate strategy and digital experience at Publicis.

Why should marketers care? Marketers should be invested in an internet where AI-generated content or content of an unknown origin can be identified to protect their brand.

Brand safety: Running ads on sites with content of unknown origin can be dangerous for brands who risk their content appearing beside mis- and disinformation.
Asset protection: Brands have an interest in establishing clarity of where assets associated with the brand come from. A nefarious version of last year’s viral image of the Pope wearing Balenciaga could prove detrimental or controversial for a brand if consumers can’t easily establish where the image came from.

These were Publicis’s motivations for backing the standard, which uses technology similar to how retailers protect credit card data using SSL encryption in order to create a record of content history. “We love that [the standard] relies on really old, boring technology that’s been around for decades,” said Buhler.

Publishers are already adopting the standard. Last month, LinkedIn adopted the C2PA standard for labeling content in an effort to increase transparency. LinkedIn joins publishers like the BBC in adopting the standard.

The standard is open source, so any publisher can adopt it using documentation available through the C2PA. The standard uses provenance data to capture and record information that demonstrates a photo, video, or audio’s origin. The person viewing content can then click on an icon to view that historical information.

Adopting a standard for content origin is also vital to the health of the internet, which marketers should be invested in. “The worst-case scenario is the internet becomes useless because it is overrun with artificial spam content generated by bots that crowds out the human voice and is indistinguishable, and we don’t know who’s real and who’s not,” warned Buhler. That future would prove disastrous for digital marketers.

Generative AI, if adopted safely, could lead to a more hopeful future. “I like to think about generative AI as democratizing creativity,” said Lansigan. He argues AI gives the tools for creation to people who don’t have technical skills or access to tools that a company like Publicis has. “That opens up an exchange of cultural ideas. It opens up an exchange of economic trade.”