AI Search
UPDATED 11:18 EDT / DECEMBER 25 2025
by Paul Gillin
The search engine optimization discipline that has guided web marketing efforts for more than two decades is now being disrupted by generative artificial intelligence systems that deliver direct answers rather than lists of links.
The good news is that much of what makes content rank highly on search engines also works with gen AI, but marketers need to make some critical modifications to optimize their visibility on ChatGPT, Gemini and others, according to one AI engine optimization expert.
The changes have been apparent since May 2024, when Google LLC’s AI Overviews summaries began appearing atop search results. They now show up on roughly 30% of U.S. searches, triggering double-digit declines in click-through rates for many websites, according to Kevin Roy, chief executive of GreenBanana SEO, the business name of Bayline Partners LLC.
TheCUBE Research, a SiliconANGLE sister company, expects that most buyers will prefer AI assistants over search engines for research and vendor comparisons by next year. The firm predicts that AI will account for over 70% of business-to-business software research by 2030.
“This decline is caused by zero-click behavior as AI systems generate and provide answers directly, bypassing websites entirely,” wrote Scott Hebner, principal AI analyst at theCUBE Research. “Traditional SEO strategies are losing visibility and control to AI engines that determine which brands to feature in their synthesized responses.”
Gen AI platforms are becoming primary discovery tools, particularly for research-oriented and professional queries. One of their biggest appeals is time savings: Instead of presenting links and requiring the user to navigate to relevant content, gen AI engines attempt to answer the question directly, often without an associated link.
Search engines determine relevance and authority based on the volume and quality of links pointing to a webpage. In contrast, generative AI engines “learn from structured data, citations and entity relationships,” Roy said.
He cited research that found the average domain age of sources referenced by ChatGPT is 17 years, suggesting that AI systems strongly favor established, consistent entities over newer or poorly structured sites. Companies that optimize solely for link-based rankings risk becoming invisible in AI-generated answers, even if their search visibility remains strong.
In effect, both search and generative AI engines look for the same qualities but do it in different ways. Rather than emphasizing keywords and backlinks, AI engines favor entity recognition, Don Dodds, founder of M16 Marketing LLC, wrote in a Forbes article in September.
“Entity optimization focuses on machine-readable identity, semantic connections across authoritative platforms and consistent brand context,” Dodds wrote. “Search and AI systems increasingly prioritize entities like people, places and organizations over strings of text.
In other words, visibility increasingly depends on how well an AI system knows what a brand is, what it does, and why it should be cited. “They’re trying to understand holistically what you are, and if you’re important writing about a subject,” Roy said.
AEO has two closely related components. The first concerns how content is structured. AI systems typically return concise, synthesized answers, which makes clearly organized source material essential. Short questions and answers, frequently asked question-style formatting and comprehensive topical coverage help models extract usable information quickly. FAQs have long been a staple of SEO; now they’re becoming table stakes, he said.
The second component is authority, which goes beyond individual pages. AI systems look for signals that other credible sources recognize and describe the same entity consistently. Structured data, particularly schema markup, plays a central role.
Schemas should use standardized JavaScript Object Notation formatting to identify elements such as authors, organizations, products and publication dates to make it easier for AI systems to interpret content without ambiguity.
“Schema used to be a nice-to-have,” Roy said. “Now it’s almost a have-to-have.”
Authorship is another increasingly important factor. Roy recommends attaching every piece of content to a real, verifiable author and creating dedicated author profile pages that link to high-authority external sources, such as professional profiles, media appearances and industry publications. This approach, called entity stacking, helps AI systems confirm that content is associated with a credible individual rather than an anonymous source.
Roy’s broader framework, which he calls Entity Authority Engineering, helps organizations “train” AI systems to recognize and trust their brand. Key elements include maintaining consistent structured data across the web, mapping brand mentions and citations across authoritative sources and testing how content performs across multiple AI models rather than optimizing for a single platform.
AEO isn’t the end of SEO, Roy asserted, but a clear dividing line is emerging. “The winners will be the people who aren’t ignoring this,” Roy said. “They’re focusing on structure, schema and real authority. The shortcuts don’t work anymore.”
AI engine optimization doesn’t replace SEO fundamentals, but it does redefine the tactics. The goal is no longer just to rank but to be recognized as an entity worth citing when someone asks a question.
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.
Anthropic to launch new ‘Claude Mythos’ model with advanced reasoning features
The ‘wild, wild west’ of agentic security: Insights from theCUBE’s RSAC closing analysis
From cloud native to AI native: The role of context density
AI agents are about to overtake cybersecurity — for better, or worse?
Oracle’s new AI bet: Make the AI database the center of agentic workloads
Cisco’s warning on agentic AI: One wrong action could cause irreversible damage to the enterprise
Anthropic to launch new ‘Claude Mythos’ model with advanced reasoning features
AI – BY MARIA DEUTSCHER . 29 MINS AGO
The ‘wild, wild west’ of agentic security: Insights from theCUBE’s RSAC closing analysis
SECURITY – BY EMILE LOUW . 3 HOURS AGO
From cloud native to AI native: The role of context density
CLOUD – BY GUEST AUTHOR . 6 HOURS AGO
AI agents are about to overtake cybersecurity — for better, or worse?
SECURITY – BY ROBERT HOF . 6 HOURS AGO
Oracle’s new AI bet: Make the AI database the center of agentic workloads
AI – BY JOHN FURRIER . 6 HOURS AGO
Cisco’s warning on agentic AI: One wrong action could cause irreversible damage to the enterprise
SECURITY – BY RYAN STEVENS . 11 HOURS AGO
Forgot Password?
Like Free Content? Subscribe to follow.
