• by Laurie Sullivan , Staff Writer, 6 hours ago

As AI agents begin to request content directly, publishers and platforms will need systems that expose enough metadata for discovery while still protecting paid content.

It’s a challenge that engineers at JWX are trying to work through, John Nardone, CEO of the company, told MediaPost.

Its "JWX Content Hub," a unified platform for publishers, announced on Monday monetization of video from one library to use across many media channels. The New York Post and McClatchy are two publishers that have been testing the platform.

Google, Cloudflare and others are building payment infrastructures that could allow AI agents to pay publishers for content, while human users continue to “pay” through attention and engagement. These are some of the ways that agentic continues to reshape access to content and ad economics.  

Search has become a thing of the past when it comes to finding specific content, replaced by artificial intelligence (AI) and agentic technologies.  

AI can generate a written article taken from the JWX Content Hub and turn it into a video. Today it uses clips of images.

The AI reads the article and interprets the content, then looks for images to create a visual representation and creates the video with human oversight through a team of video editors that support publishers if needed.

Aug X Labs -- the startup behind Augie Studio, the AI-assisted “Commercial Video Studio” -- created the tool before JWX acquired it earlier this year.

The acquisition supported JWX’s vision to expand its publisher platform beyond video infrastructure.

The deal now helps media companies change existing content into multiple formats, distribute that content efficiently across fragmented platforms, and drive deeper audience engagement. 

Nardone said that in the not-too-distant future, sophisticated technology people will create their own agents that can bring back publisher content to them, adding that JWX is actively working to determine what that technology looks like, as the percentage of agentic-related content requests continue to rise.

“Video presents a unique challenge for agents, which you cannot crawl the page and get a video because it sits in a content management system,” he said. “In this case, the agent must ask for the content, but how does it know what to ask for and how does it know what’s in the content library.”

The publisher must expose enough content in the library so the agent knows what is in the video. But the trick is determining a new way to do this without requiring the agent to purchase it before it knows it needs it.

“The article might cost three cents,” Nardone said. “The publisher will need an interface to manage the interaction, which is coming.”

Remember the RSS (really simple syndication) feed, which allowed users to access updates to websites and applications in a computer-readable format. Subscribing to RSS feeds meant the publisher would push real time content updates to the subscriber.

AI replaces that concept. In the model for the companies building a micropayment infrastructure -- including Google and Cloudflare -- publishers will ask agents to pay to access their content, while humans will make the payment with their attention.

Nadav Perry, newly appointed CRO for the Americas at Ogury, has Google Gemini read his agenda every morning as he commutes to work in New York to prepare for meetings in about 10 minutes because the AI assistant digests “tons of data” and summarizes it.

Perry spent nearly 10 years at Google, and became the chairman of the Xoogler IL Community after leaving the company in 2019 to join Taboola, where he remained until joining Ogury five days ago.

Perry said the community is full of former Google employees, now founders and entrepreneurs, working in all types of media as the advertising and technology industries explode into areas it has yet to experience.

It’s no secret that natural language has replaced structured queries in search.

“It’s mind blowing what has happened to search and the capabilities that Google Gemini and Claude and ChatGPT are bringing to life, but from an advertiser’s view it brings a new set of challenges,” he said -- such as showing the brand in an environment where answers already live -- for example, in large language model (LLM) native advertising.

Perry sees promise for Ogury, a global ad-tech company specializing in audience intelligence, to solve an ad waste challenge and make running advertisements across multiple platforms more affordable using artificial intelligence. The pace of change is exponential, he said.

This is something that Suganthan Mohanadasan, co-founder of Snippet Digital, an AI SEO agency, pointed out in a recent post.

On August 16, ChatGPT ran a web search using the tool JSON, and just four days later it ran the same web query using a compact query language of its own. The original language used to find Mohanadasan’s first query had vanished completely.

“A tool call is now a block of lines, one search per line, fields separated by pipes,” he wrote. “The pattern that repeats is a call type, the query, a number, and sometimes a domain.”

Mohanadasan analyzed how a search engine determines the length of the text snippets extracts from web pages based on the user's specific search intent, so he suggests beginning each section with a direct answer to ensure that it makes sense to users and accepts the click.