Attribution in an AI world: a challenge and an opening for news publishers

As AI reshapes news discovery, publishers face the decline of the link economy and the rise of traceability as a new foundation.

In 2025, news is summarized, transformed, and increasingly surfaced by AI agents that operate with little regard for attribution. For publishers, that shift represents a real existential challenge. But for those correctly prepared, with the right partners, it’s also a powerful opportunity.

The erosion of the link economy

As generative AI platforms become amongst primary discovery layers for content, traditional referral traffic from search continues to decline, while social media still struggles to find a steady pace (watching at BlueSky or Reddit for instance). 

The Reuters Institute's 2025 Digital News Report highlights that fewer than 25% of under-35s now access news through publisher homepages, and more than half can’t reliably identify which stories were AI-generated. The shift from search-driven clicks to AI-summarized answers is breaking the link economy, where clicks meant revenue and reach.

Traceability: more than metadata

What fills that void is the need for immediate traceability. In this new environment, trust isn’t established by design or reputation alone, it must be embedded, readable by machines, and resilient through transformation. That means a persistent metadata that travels with the content across platforms and formats, usage tracking systems that show who’s consuming, modifying, or monetizing the work and clear indicators of AI involvement and original authorship. 

Tools like content credentials and digital watermarking play a role, , and industry-led initiatives such as the C2PA (Coalition for Content Provenance and Authenticity) are helping establish shared standards for content integrity. C2PA’s open framework allows media organizations to embed tamper-evident metadata, making it easier to verify who created what, when, and how.

But traceability is bigger than any single solution. It’s the infrastructure needed for fair licensing, ethical AI training, and public trust.

Distribution without disappearance

Channels, old and new, are recognizing this shift too. Recent introduction to "pay-per-crawl" functionalities, will soon give content owners more control over how automated agents access their material. Similarly, edge technologies are emerging to help publishers monetize or gate AI bot traffic, turning what was once invisible usage into measurable value.

The goal isn't to block AI outright, but to ensure that journalism remains visible, valued, and credited, even when it's being ingested by a model, not read by a human.

Format fluidity, fixed Identity

At Nordot, we help publishers scale smartly: reaching more platforms, audiences, and formats, while preserving control, identity, and revenue potential. But even as we enable transformation, we believe in preserving the origin as format agility must come with fixed identity. For the publisher, for the channel its content is distributed on, and for the readers, first and foremost.

That’s why attaching source and licensing metadata to all content variants, tracking real-time usage across syndication environments and helping publishers maintain editorial control, even in downstream applications is at the very foundation of our daily work.

Our recent work with The Brighter Side of News showed how a lean operation can grow its audience by a lot without changing what it publishes, just how it distributes and tracks it. That’s the kind of scale traceability enables.

A shared responsibility

As AI reshapes discovery, traceability must become a shared standard across media ecosystems. This includes publishers and platforms aligning on metadata protocols, AI companies respecting structured content rights and licensing, and developers integrating provenance directly into media workflows.

Creating future-proof systems that make journalism identifiable, creditable, and economically sustainable in an AI-first world must be at the heart of our ecosystem. It’s how we build trust, protect content, and keep journalism visible where it matters most.

Bertrand de Volontat

Bertrand de Volontat is VP of EMEA at Nordot, where he helps publishers make the most of their content across platforms and borders. With a background in both journalism and media business, he led editorial operations at upday and launched several digital media ventures in the past.

Back to Publisher's Playbook

Curious to Learn More?

Let's Connect!