admin
The rapid evolution of artificial intelligence and the proliferation of Large Language Models (LLMs) have fundamentally altered the landscape of digital information retrieval, rendering traditional communications playbooks increasingly obsolete. As search engines transition into "answer engines," the methodologies used by public relations and marketing professionals to ensure brand visibility are undergoing a radical shift. This transformation was the focal point of a recent intensive workshop led by industry leaders Sukhi Sahni, a fractional CMO and advisor; Gini Dietrich, founder of Spin Sucks and creator of the PESO Model; and Dr. Sarab Kochhar, senior communications officer at the Gates Foundation. The consensus among these experts is clear: the rules of engagement have changed, and the integration of Paid, Earned, Shared, and Owned media must now be viewed through the lens of machine readability and algorithmic trust.
The Convergence of AI Visibility and Strategic Communications
For decades, the public relations industry relied on a predictable cycle of media outreach, press release distribution, and earned media placements. However, the advent of generative AI tools such as ChatGPT, Claude, Perplexity, and Gemini has introduced a "zero-click" environment. According to recent industry data, nearly 60% of digital searches now conclude without a user clicking through to a third-party website. Instead, users receive synthesized answers directly from the AI interface. For communications professionals, this means that "ranking" on the first page of Google is no longer the sole metric of success; the new objective is "citation" within the AI’s generative response.
During the workshop hosted by Ragan, the discussion moved beyond the basic utility of AI tools and focused on the structural failures of current communication strategies. Many organizations continue to operate in silos, where earned media teams do not coordinate with owned media creators, and shared media channels function independently of the broader brand narrative. This fragmentation creates a "visibility gap" that AI models struggle to bridge. When information is scattered and inconsistent, LLMs are less likely to cite the organization as a primary or authoritative source.
From AEO to GEO: A New Technical Frontier
A critical distinction highlighted during the session was the transition from Answer Engine Optimization (AEO) to Generative Engine Optimization (GEO). While AEO has been a component of SEO for several years—focusing on featured snippets, "People Also Ask" boxes, and voice search—GEO represents a more sophisticated challenge. GEO involves optimizing content specifically to be ingested, synthesized, and cited by generative AI.
The playbooks for these two disciplines differ significantly. While AEO relies heavily on schema markup and concise "FAQ-style" formatting, GEO requires a deeper level of semantic authority. To be visible in a generative engine, a brand’s content must not only answer a question but must do so with a level of verifiable expertise that the model’s training data recognizes as high-quality. This shift necessitates a move away from vague corporate storytelling and toward data-driven, evidence-based content.
The Criticality of Original Data Interpretation
One of the most common concerns raised by communications professionals is the perceived lack of original data within their organizations. However, the workshop leaders argued that owning the raw data is less important than owning the interpretation of that data. AI models are programmed to look for specific, citable claims rather than anecdotes.
A landmark study conducted by researchers at Princeton University, the Allen Institute for AI, and IIT Delhi demonstrated the tangible impact of content structure on AI visibility. The study found that the inclusion of statistics, direct citations, and authoritative quotations can increase a brand’s visibility in AI-generated responses by as much as 40%. For example, a university claiming "high student success rates" offers little value to an LLM. Conversely, an institution stating that "80% of graduate students find employment in their field within six months of graduation" provides a discrete, citable fact that the AI can use to build its response.
To address this, organizations are encouraged to audit their "expertise" rather than just their "content." This involves identifying internal experts, formalizing their insights into white papers or technical blogs on owned domains, and ensuring that every claim made in a press release is backed by a permanent, linkable source on the organization’s website.
Wikipedia and Reddit: The New Infrastructure of Trust
Perhaps the most surprising revelation for many attendees was the dominant role of community-driven platforms in shaping AI outputs. According to the 5W AI Platform Citation Source Index 2026, Wikipedia remains a cornerstone of the AI information ecosystem, accounting for 26% to 48% of the citations generated by ChatGPT. Despite its importance, many communications teams treat Wikipedia as an afterthought or a platform to be avoided due to its strict "Conflict of Interest" (COI) guidelines.
The workshop emphasized that a "Wikipedia vacancy"—the absence of a well-maintained, neutrally written, and properly sourced entry—is a strategic liability. Because AI models use Wikipedia as a primary source of ground truth, an organization without a presence there is effectively invisible to the AI’s foundational knowledge base. Establishing a professional relationship with the Wikipedia community, understanding editing policies, and ensuring that earned media coverage is used to support Wikipedia citations is now a core requirement of the PESO model.
Similarly, Reddit has emerged as a primary source for "human-centric" queries. The 5W analysis identified Reddit as the top-cited source across major AI engines, appearing in approximately 40% of all citations. This represents "Shared" media performing the work of "Earned" media. AI models prioritize Reddit because it contains high-density, conversational data that reflects real-world sentiment. Communications teams are now tasked with monitoring Reddit not just for sentiment analysis, but to understand the specific language and questions their audience is using, which can then be mirrored in the organization’s owned content to improve AI alignment.
Reimagining the PESO Model for an AI-First Era
The PESO Model, originally developed by Gini Dietrich, has long served as a framework for integrating Paid, Earned, Shared, and Owned media. In the context of AI, this model must function as a singular, connected system rather than a collection of tactics.
- Owned Media: This serves as the "source of truth." In an AI-driven world, owned media must be structured with clear headers, bulleted lists, and robust internal linking to make it easily "crawlable" for AI bots.
- Earned Media: Beyond the initial PR value, earned media now serves as third-party validation for AI. When a reputable news outlet cites a brand’s data, it reinforces the authority of the brand’s owned content in the eyes of the LLM.
- Shared Media: Social platforms like LinkedIn and Reddit are no longer just for engagement; they are signals of relevance. However, the workshop noted that technical nuances, such as LinkedIn’s suppression of "link in comments" posts, require teams to stay agile in their distribution tactics.
- Paid Media: Paid strategies are increasingly used to amplify the reach of owned "authority" content, ensuring that it gains enough traction to be picked up by shared and earned channels, eventually feeding back into the AI’s training data.
Executive Thought Leadership and the Human Element
Despite the focus on machine readability, the human element remains a decisive factor in B2B communications. The 2025 Edelman-LinkedIn B2B Thought Leadership Report found that 95% of "hidden buyers"—those who research products and services before ever contacting a salesperson—state that strong thought leadership makes them more receptive to a brand’s outreach.
The workshop addressed the common challenge of reluctant executives who may not wish to be active on social media. The solution, according to the experts, is consistency over seniority. Identifying subject matter experts within the organization who are already engaged in industry conversations and helping them publish consistently can be more effective than a ghost-written, infrequent post from a CEO. This "Shared" media presence provides the conversational data points that AI models crave when synthesizing answers about industry trends and leadership.
Analysis of Broader Implications and Long-term Impact
The shift toward AI-centric communications is not a temporary trend but a fundamental restructuring of the information economy. Organizations that fail to adapt their playbooks within the next 12 months risk being "erased" from the digital record. When an AI model synthesizes an answer about a specific industry or category, it relies on the most accessible, authoritative, and consistent data available. If an organization’s content is not structured to meet these criteria, the AI will simply cite a competitor who has prioritized GEO.
Furthermore, the "crisis management" utility of this approach cannot be overstated. In a breaking news scenario, journalists frequently turn to AI to gather background information. If an organization has not built a foundation of well-sourced, current, and consistent owned content, the AI will pull from whatever earned media exists—which may include outdated coverage, critic commentary, or inaccurate third-party reports.
The ultimate takeaway from the Sahni, Dietrich, and Kochhar workshop is that the "wait and see" approach to AI is no longer viable. The playbook for the next decade of communications is being written now, and it is being written by those who understand that the bridge between a brand and its audience is no longer just a screen, but a complex algorithmic interpreter. Success in this new era requires a commitment to structural integrity, data-backed authority, and the seamless integration of all media types into a single, AI-readable ecosystem. Organizations that master this transition will define their categories for years to come; those that do not will find themselves struggling to correct a narrative that was written without their input.
