The morning ritual for Michelle Seo, a public relations consultant based in San Diego, now includes a briefing from Chattiana, her AI agent. This daily digest, curated by her Muse AI, not only outlines Seo’s agenda and to-do list but also extracts relevant water-cooler topics from her newsletters, complete with source links. "I’m kind of challenging myself as an introvert to start more conversations," Seo shares with WIRED. This AI-generated briefing provides her with a snapshot of the daily news cycle and potential conversation starters for virtual meetings, a far cry from traditional news consumption.
Seo’s experience is indicative of a burgeoning trend: the increasing reliance on artificial intelligence for news aggregation and consumption. This shift is so profound that a segment of the media industry is beginning to tailor its content and distribution strategies not for human readers, but for AI agents themselves. This innovative approach manifests in various forms, from optimizing content feeds specifically for AI consumption to structuring articles in ways that facilitate personalized summaries and analyses by these intelligent systems.
The Shifting Media Landscape: Why Publishers Are Courting AI
The rationale behind this strategic pivot is rooted in a perceived erosion of traditional digital media consumption patterns. Jacquelyn Cameron, Chief Revenue Officer at Axios, articulates this evolving ecosystem: "We believe that the media ecosystem is changing. We see this with the open web deteriorating. We see that search is shrinking. We see that hyper-personalization is surging, that LLMs are becoming more important in the way that people are consuming their news." This observation is supported by data indicating a decline in organic search traffic for many publishers and a corresponding rise in traffic driven by large language models (LLMs). The Pew Research Center has reported a significant increase in adults using AI tools for information retrieval, underscoring the growing influence of AI in how people access knowledge.
Major media organizations have responded to this paradigm shift with a spectrum of strategies. Many, including Condé Nast, the parent company of WIRED, have entered into agreements with AI platforms to ensure their content is surfaced by these emerging technologies. This includes grappling with the concept of "generative engine optimization" (GEO), a strategy analogous to Search Engine Optimization (SEO) but tailored for AI. The goal is to ensure that articles are readily discoverable and comprehensible when AI models are prompted for information. Last year, AIOSEO, a popular SEO plugin for WordPress, introduced LLM.txt file generation, designed to assist AI in discovering and interpreting website content, a move that signals a proactive adaptation to AI’s crawling and comprehension capabilities.
However, this adaptation isn’t without its conflicts. The New York Times notably filed a copyright infringement lawsuit against OpenAI and Microsoft in 2023, alleging that the companies had illegally trained their AI models on millions of copyrighted articles without permission or compensation. This landmark case, still ongoing, highlights the complex legal and ethical terrain publishers are navigating as their content becomes a foundational element for AI development.
Axios’s Bold Move: Direct Feeds for AI Agents
Amidst these evolving dynamics, Axios has taken a particularly forward-thinking approach, aiming to serve AI agents directly. At the Digiday Publishing Summit in September, the company announced the impending launch of Axios Direct, a series of data feeds specifically engineered for AI agents. "We do believe that a significant portion of news consumption will move into AI systems, LLMs, and devices," Cameron stated. She envisions these feeds as a streamlined conduit for AI to access Axios content, prioritizing structured data over narrative flow, akin to a Bloomberg terminal rather than a traditional article.
The initial phase of Axios Direct targets corporate clients, reflecting a strategic move to capitalize on business intelligence needs. The first feed, launched in the fall, grants AI agents of Axios Pro Deals subscribers access to merger and acquisition data. A second feed, slated for release next year, will serve as a training dataset for in-house large language models at client organizations. This B2B focus is pragmatic, as corporate entities often possess larger budgets and a greater tolerance for experimental technologies, especially if they offer a competitive edge.
The roadmap, however, includes a consumer-facing element. Next year, Axios plans to launch a third Axios Direct feed designed for individual subscribers, enabling their personal AI agents to receive tailored reporting. These feeds will adopt a format resembling RSS feeds: functional and standardized lines of text, devoid of the aesthetic considerations intended for human readers. This approach signifies a fundamental re-evaluation of content presentation, prioritizing machine readability over human visual appeal.
Every’s Innovative Approach: AI as an Extension of the Reader
While Axios optimizes its content for AI extraction, Every, a six-year-old tech media and software company, has embraced a different philosophy. Having secured a $2 million funding round last year, which included investment from LinkedIn founder Reid Hoffman, Every positions AI as an integral extension of its subscribers’ own capabilities.
Dan Shipper, founder of Every, explains that their content is often crafted with AI interaction in mind. Writers frequently embed prompts directly into articles, empowering readers to experiment with concepts and test hypotheses using AI. For instance, an article discussing a new AI model might include a series of copy-pasteable prompts that allow readers to observe how the model responds to specific questions and evaluates inputs. This pedagogical approach transforms passive reading into an active, AI-assisted learning experience.
Furthermore, certain article formats, such as comprehensive guides, feature interactive buttons allowing readers to engage with AI platforms like Claude or ChatGPT. Clicking these buttons initiates a pre-defined prompt, such as "Start with a tight summary: one paragraph, bulleted. Then offer to go deeper." This feature enables writers to pack their stories with extensive detail, confident that AI agents will efficiently extract the most pertinent information for individual users, bypassing the limitations of human attention spans.
Two years ago, Every pioneered its "extendable" article format, which pairs long-form journalism with a comprehensive database of supporting materials, including interview recordings, original news articles, and videos. Subscribers can query a chatbot trained on this database to access additional context and details, creating a dynamic and deeply explorable reading experience. Most recently, the publication released an 8,000-word article on Codex, which subscribers can consume in an even wider array of formats. Their AI agents can leverage the associated database to generate personalized versions of the content—whether an article, podcast, or video—tailored to specific user preferences. One subscriber might receive a bulleted summary, while another could request a short film compiling key interview segments.
"As a writer, you have to make choices about what goes into the story you’re telling," Shipper reflects. "And there’s always a question about, are those the right choices? Or, are those the right choices for a particular reader? With AI, journalists no longer choose." This liberation from editorial constraints, Shipper suggests, will lead to novel and unexpected applications of journalistic content.
The Monetization Maze: Can AI Agents Become a Revenue Stream?
As media outlets increasingly view AI agents as a distinct audience, the crucial question of monetization arises. Rasmus Kleis Nielsen, a professor of communication at the University of Copenhagen, characterizes the market for publishing to AI agents as "emerging" but "extremely uneven." He points to significant unanswered questions regarding pricing models and the long-term interaction between AI agents and published content.
Despite these uncertainties, executives at Axios remain optimistic. Allison Murphy, Axios’s Chief Operating Officer, highlights that the publisher’s concise, structured writing style naturally lends itself to AI consumption. "Time is money for crawlers," Murphy explains. "They don’t want to have to go three quarters down the page to find the main point. So, that’s something that’s already benefiting us."
The financial benefit, however, is not solely tied to direct traffic. Nielsen suggests that even if AI agents do not drive direct visits to a publisher’s homepage, the brand visibility generated through AI-powered summaries and responses can be valuable. This mirrors the dynamic seen with content optimized for social media platforms like TikTok and Instagram, where brand presence can be cultivated even if user engagement does not translate into website traffic.
Publishers generally face challenges in monetizing content that can be easily synthesized by chatbots. The premium is increasingly placed on information that is unique, proprietary, or offers an exclusive advantage. At Every, this translates to in-depth guidance on AI utilization, while Axios Pro Deals offers rapid access to market-moving financial information. An all-access Axios Pro Deals subscription costs $2,499 annually, whereas an Every subscription is priced at $288 per year.
Shipper views the $288 subscription not just for the human reader but also for their AI agents. "I may not want to read all of your stuff every month," he concedes. "But if my subscription can include my agent, I may want my agent to know it so that it can tell me, ‘Hey, this thing is important.’" This perspective reframes subscription models to encompass both human and artificial intelligence as consumers.
Murphy emphasizes Axios’s stance of acceptance, acknowledging that AI is already a significant tool for information consumption and will likely become even more so. "I also try to make it a little less existentially terrifying by thinking, this isn’t the first time this has happened in media," Murphy notes, drawing parallels to the adoption of social media, short-form video, and podcasts as news delivery mechanisms. Ultimately, Murphy concludes, "It’s not really AI as the audience. It’s AI as a tool that helps us reach our audience." This framing positions AI not as a replacement for human readership but as a powerful new channel for engaging with them.
