What old patents reveal about AI search

Every time a new large language model (LLM) drops or Google tweaks an AI Overview, the SEO industry loses its mind. We develop this weird collective amnesia, scrambling to optimize for features that were actually mapped out in patent offices 10 years ago. We’re so obsessed with the now and the next that we’ve stopped looking at the blueprints.

If you want to survive 2026, stop trying to be a futurist. Instead, be an archaeologist.

To actually deliver for our clients, we need a research framework that isn’t just reactive. It has to be a balance: Look back at the foundational patents to understand the rules, and look ahead to see how AI is finally being given the muscle to enforce them.

The archaeology of SEO

There’s a massive misconception that to understand AI search, you need to be a prompt engineer or read every new research paper from OpenAI. You don’t.  The logic governing today’s magic is often math that was written a decade ago.

We can’t talk about patent research without honoring the late, great Bill Slawski. For 20 years, he was the SEO industry’s archaeologist. While everyone else was arguing about keyword density, he was reading dry, technical filings to predict exactly where we’re standing right now.

History proves his method worked.

The algorithm isn’t magic. It’s math. When a new feature drops today, the engineering blueprints were likely filed between 2007 and 2016. If you want to win, go read the old stuff.

Dig deeper: The origins of SEO and what they mean for GEO and AIO

Strategy vs. mechanics: From ‘strings’ to ‘verified things’

Don’t get buried in buzzwords. Categorize your learning into two buckets: ”strategy” or ”mechanic.”

For years, the industry talked about moving from strings to things (entities). But in 2026, that’s just the baseline. We’ve moved from strings to verifiable things. An entity is worthless if the AI can’t prove it’s real.

Think of it like building a house:

• Semantic SEO is the architecture: It’s the vision. It’s making sure the meaning of your site actually matches what the user is looking for.
• Entity SEO is the bricklaying: It’s using distinct nouns to build that vision so a machine can parse it.
• Verification is the mortgage: This is the part most people miss. It’s turning those entities into findable, provable facts connected to a verified human. If you aren’t connecting your content to a provable human expert, you’re just adding to the noise.

AEO vs. GEO: Let’s stop using these interchangeably

The industry often uses AEO and GEO synonymously, but they require different content structures and serve different objectives.

Answer engine optimization (AEO)

AEO is for the “direct answer.” Think Siri, Alexa, or that single snippet at the top of the page. It’s binary. It’s rooted in those 2006 fact repository patents.

You need ”confidence anchors.” These are unnuanced, structured facts. The engine isn’t “thinking,” it’s fetching. If your fact isn’t provable and anchored to a verified source, the engine won’t risk a hallucination by citing you.

Generative engine optimization (GEO)

GEO is for the “synthesis.” This is Gemini or ChatGPT search explaining how something works. It was formally defined by researchers at Princeton and Georgia Tech in 2023.

You need information gain. These engines don’t just want a fact; they want to see how Concept A affects Concept B. They’re looking for relationships and unique perspectives.

In short, AEO is about being the fact. GEO is about being the authority that the AI trusts to explain those facts.

Dig deeper: SEO, GEO, or ASO? What to call the new era of brand visibility in AI [Research]

The trap of forward-projecting: Why the ‘basics’ are still the ‘floor’

There’s a danger in becoming an SEO time traveler. If you spend all your time in the patent archives or stress-testing GEO relationships, you might forget that the AI still has to reach your content.

You can have the most verified, E-E-A-T-heavy content in the world, but if your site’s technical health is a mess, the confidence anchors will never weigh in.

The persistence of technical debt

Basic SEO requirements haven’t changed. The tolerance for ignoring them has simply disappeared.

• Crawl budget and efficiency: If your site is bloated with zombie pages or redirect loops, you’re wasting the crawler’s time. LLMs aren’t just looking for content. They’re looking for the cleanest path to a fact.
• Core Web Vitals (CWV): More than a ranking factor, it’s a user-utility requirement. If your site doesn’t load instantly, the AI won’t recommend it as a source in a GEO overview.

The headless promise (and reality)

Many of the frustrating technical SEO issues we’ve fought for years — like bloated JavaScript and poor Largest Contentful Paint (LCP) — are finally being solved by headless/composable architectures. By decoupling the front end from the back end, we can deliver the raw, lightning-fast data that answer engines crave while maintaining a high-end experience for humans.

But headless isn’t a “get out of SEO jail free” card.  It solves the speed problem, but it introduces new risks around dynamic rendering and metadata delivery.

Whether you’re on a 20-year-old CMS or a cutting-edge headless build, the today requirements are non-negotiable:

• Clean URL structures: If the AI can’t deduce the hierarchy from the URL, you’ve already lost the semantic battle
• Internal linking (the nervous system): This is how you prove relationships between entities. If your internal linking is broken, your synthesis logic doesn’t exist.
• Indexability: If the bot is blocked by a poorly configured robots.txt or a noindex tag left over from staging, the most brilliant “verified human” insights in the world are invisible

You don’t get to play in the frontier of AEO and GEO until you’ve mastered the floor of technical SEO. Don’t let the shiny new objects make you forget the shovel work.

Dig deeper: Thriving in AI search starts with SEO fundamentals

The SEO time traveler checklist

Phase 1: The archive

• The Slawski deep dive: Stop reading the latest “AI is changing everything” blog posts for five minutes. Go back to the SEO by the Sea archives. Search for Slawski’s analysis on the Knowledge Graph or the user context. You’ll see the 2026 roadmap hidden in plain sight.
• The E-E-A-T math audit: Check your assets against Patent 2015/0331866. Are you actually providing the contribution metrics (such as verifiable reviews) that the patent specifically asks for?

Phase 2: The laboratory

• The verification pivot: Audit your entities. Are they just names on a page? Link them to a verified LinkedIn profile or a Knowledge Panel. If it’s not verified, it’s not an entity, it’s just a string of text.
• Schema stress testing: Don’t just use a plugin and walk away. Experiment with nesting. Try nesting a Person inside a Service as the provider. It works — I’ve seen it trigger rich results when nothing else did.

Phase 3: The frontier

• The confidence anchor audit: Look at your top pages. Does every topic have a clear definition? [Entity] is [attribute]. If you’re being vague, you’re invisible to AEO.
• The synthesis test: This is a quick one. Paste your article into an LLM and ask it to explain the relationship between your two main topics using only your text. If it has to go to the web to find the answer, you haven’t built the relationship well enough for GEO.

The synthesis: Becoming the architect

The SEO time traveler isn’t looking back because they’re nostalgic. They’re looking back because they want the blueprint. When you realize AEO is just the modern enforcement of a 20-year-old patent and GEO is just the evolution of semantic relationships, the chaos of AI updates disappears.

Stop optimizing for strings. Start optimizing for verified facts. Give the engine a fact it can’t doubt, connected to a person it trusts, and a relationship it can’t ignore.

The future of search wasn’t written this morning — it was written years ago. You just have to be the one to actually build it.

Dig deeper: The future of SEO: Why optimization still matters, whatever you call it

References and further reading 

On the evolution of fact-based search (AEO foundations)

On generative engine optimization (GEO foundations)

• The GEO framework: Aggarwal, V., et al. (2023). GEO: Generative Engine Optimization. Princeton University, Georgia Institute of Technology, and the Allen Institute for AI. The definitive study on how LLMs cite and prioritize authoritative sources. 
• The Slawski legacy: Slawski, B. (Various). SEO by the Sea Archives. For historical context on Agent Rank, phrase-based indexing, and entity metrics.