An analysis of who controls AI answers in the insurance sector, identifying the key sources, publishers, platforms, and regulators influencing AI retrieval and citation patterns. It highlights the dominance of consumer-finance publishers, the role of government sites, and the limited impact of carrier-owned content.
The structural shift is complete. What actually changed, what it means for every brand and communications team, and what the next 18 months look like.
The first half of 2026 established GEO as a defined discipline. Five things that happened, five things that didn't, and five things that will determine where this goes.
Reddit. Wikipedia. .gov sources. The category-native trade publication. The named individual practitioner. Five source types appear in the top-10 cited sources for virtually every industry vertical — and what each means for a GEO program.
Fifteen verticals mapped. Five patterns repeat without exception: category-native beats legacy, .gov anchors the factual floor, Reddit dominates experience queries, named practitioners out-cite firms, and revenue leadership ≠ Citation Share leadership.
G2 and Gartner own buyer-decision prompts. Stack Overflow and HN own technical. Vendor docs are quietly running the table.
Claude is the answer engine that most rewards analytical depth, named authorship, and long-form editorial. Here is the complete guide to earning citation in Claude's answers.
AI engines don't have opinions. They synthesize from sources. Understanding exactly which source signals carry the most weight — and why your website matters less than you think — is the foundational knowledge for any GEO program.
Lanes 9 and 10 of the AI Communications 100: the investors shaping the AI development narrative, and the foundations whose accumulated content built the substrate AI engines train on.
The source map for EdTech AI answers. Government validation sources anchor the factual floor. Category-native trade press sets the editorial tier. Named researchers own the expertise layer.