BUY INNOVATIONQ+

250,000 Patents and Counting: Why Patent Thickets Are Now an R&D Problem

by | May 28, 2026 | Ideation, Evaluation, & Collaboration, Artificial Intelligence, IP Intelligence, News

Navigating AI assisted ideation

A modern smartphone is reportedly covered by as many as 250,000 active patents. A quarter of a million overlapping protections, touching everything from the antenna and battery chemistry to the way a finger swipes across a screen, sit between any new device and the consumer who buys it. That is not a competitive moat. That is a minefield. And it has a name: a patent thicket.

If you’ve spent any time on legal-tech blogs or in the R&D patent space, you’ve heard the term. But the substance behind the trend is not new, and it’s not just a buzz word in the industry. It’s a strategic risk that increasingly determines which products succeed, which launch on time and hold market share, and which R&D budgets earn their return. The teams treating thicket analysis as a litigation problem are already behind. The teams treating it as an R&D and strategy problem are pulling ahead.

What a Patent Thicket Actually Is

The cleanest definition still belongs to UC Berkeley economist Carl Shapiro, who coined the modern usage in his 2001 paper Navigating the Patent Thicket: Cross Licenses, Patent Pools, and Standard-Setting. Shapiro described a patent thicket as “a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialize new technology.”

That’s the picture: you have a viable invention, you have funding, you have a market, and standing between you and revenue is a tangle of patents owned by competitors, suppliers, customers, standards bodies, universities, and non-practicing entities. Each is a potential infringement claim. Each is a potential licensing negotiation. Each, if missed, is a potential lawsuit.

Thickets occur most visibly in industries where dozens or hundreds of patented technologies converge in a single product: telecom, semiconductors, software, automotive, AI, and increasingly, biologics. Where Shapiro originally framed the problem, the empirical literature has since reinforced it. Economists Noel and Schankerman found that patent thickets reduce R&D investment by raising the marginal cost of innovation, and follow-up work by Galasso and Schankerman documented measurable declines in innovation output in technologically complex sectors where ownership of patent rights is fragmented across many holders.

The Existential Issue: Cost, Time, and Foreclosed Markets

Why does this matter now? Because the cost of getting it wrong is no longer hypothetical.

According to the AIPLA Report of the Economic Survey, the median total cost of a U.S. patent infringement suit with more than $25 million at risk is roughly $5 million, and even small cases (those with under $1 million at stake) routinely run $600,000 or more to litigate. In 2024 alone, U.S. courts handed down 72 patent jury verdicts totaling over $4.19 billion in damages, with twelve individual cases exceeding $100 million each. Those numbers don’t include settlement payments, royalty stacks, injunctions, or the opportunity cost of a product that never ships.

The takeaway is uncomfortable but unavoidable: by the time you discover you’re inside someone else’s thicket, your strategic options have already collapsed. You are negotiating from the weak side of the table, with a launch date on the line and capital already sunk into product development.

This is where we’ll plant the flag: patent thicket analysis is not a litigation activity. It is a strategy and R&D activity, and it belongs at the front of the innovation lifecycle, not the back. Companies that treat freedom-to-operate (FTO) and landscape analysis as a checkbox before launch are making a category error. By then, the design has solidified, the supply chain is locked, and the patents in the way are the patents you have to deal with. Pushed upstream into ideation, prior-art search, and portfolio strategy, the same analysis becomes a creative input. You can design around problem patents, target whitespace, identify licensing partners, or make a deliberate decision to defensively publish a feature rather than file on it.

The $12.5 Billion Shield, Inside the Smartphone Wars

The most cited example is also the most instructive. Between 2010 and 2018, the global smartphone industry was the site of one of the most expensive patent campaigns in modern history. Apple and Samsung alone were litigating more than 50 cases across 10 countries at the peak of the dispute, with billions in damages claimed between them. A 2012 U.S. jury verdict awarded Apple $1.049 billion against Samsung for design and utility patent infringement, a number that bounced through appeals for years and ultimately reached the Supreme Court.

The most telling number, though, isn’t a damages award. It’s the $12.5 billion Google paid for Motorola Mobility in 2011, primarily to acquire its roughly 17,000 patents as a defensive shield against Apple and Microsoft. That is the price of entering a thicket without one of your own. When a company spends the GDP of a small nation buying patents not to use them, but to deter being sued, the patent system has stopped functioning as a pure incentive to innovate and started functioning as a tax on market entry. Every smartphone maker, and every startup hoping to become one, paid that tax.

The smartphone thicket is also, importantly, a cross-licensing thicket. Tech companies amass huge portfolios because they need to. No single firm owns all the patents required to ship a phone. The patents get pooled, traded, and licensed at scale. That dynamic is painful, but it is at least navigable. Which makes the next example all the more instructive, because pharmaceutical thickets do not work the same way at all.

Humira’s Fortress, Where Pharmaceutical Thickets Play by Different Rules

AbbVie’s blockbuster biologic Humira (adalimumab) was first approved by the FDA in 2002. Its core composition patent expired in 2016. Under normal patent dynamics, that should have opened the door to biosimilar competition and a steep price drop for a drug used to treat rheumatoid arthritis, Crohn’s disease, ulcerative colitis, and other chronic conditions.

It didn’t. Instead, AbbVie filed an estimated 247 patent applications covering Humira’s manufacturing methods, formulations, dosing regimens, and use cases, ultimately securing 132 granted patents, the vast majority issued after 2014, more than a decade after the drug was already on the market. The result was a thicket so dense that biosimilar manufacturers, facing the prospect of litigating dozens of patents individually, chose to settle and delay U.S. market entry until 2023, even as biosimilar competition arrived in Europe years earlier. Over that delay window, Humira generated more than $114 billion in cumulative U.S. revenue.

Here is where the pharmaceutical story breaks from the smartphone story. Rutgers Law professor Michael Carrier and S. Sean Tu argue in their 2024 Texas Intellectual Property Law Journal essay “Why Pharmaceutical Patent Thickets Are Unique” that pharma thickets are a fundamentally different mechanism than the cross-licensing thickets of high tech. Their central point: pharmaceutical companies already have every patent they need to bring a drug to market. They don’t need to license from competitor organizations. So when they accumulate hundreds of additional patents on a single product, those patents are not bargaining chips for cross-licensing. They are pure exclusion tools, designed to keep their competitors out.

Carrier and Tu back this up with empirical evidence on continuation patents, which are duplicative filings that cannot disclose any new matter. They found that continuation filings have risen sharply in pharma compared to high tech, and that the pharma industry litigates those continuations at a markedly higher rate. The same pattern holds for “method-of-use” patents that drug firms have long used to delay generic entry. Layered on top of the FDA’s regulatory barriers and the high concentration of the biologics market, the thicket becomes something closer to a fortress than a hedge.

The strategic lesson is sharper because of this. In tech, you can sometimes cross-license your way out of trouble. In pharma, you usually cannot. If you are a biosimilar maker, a generics manufacturer, or a smaller biotech trying to enter an established space, the thicket is the strategy you have to beat, not a thicket you can join. The company that maps the thicket first wins. AbbVie built one. Its biosimilar competitors didn’t see the full shape of it until they were already inside.

The Operational Shift: Move Thicket Analysis Upstream

The fix is not legal. It’s operational. Thicket analysis has to live inside the R&D and IP strategy workflow, not at the end of it. Three disciplines, in order:

  • Map the landscape before you scope the project. Patent landscape and competitive intelligence reports should be table stakes at the start of any meaningful R&D initiative, not a one-time deliverable, but a living view of who owns what in your technology space and how the ownership is fragmenting over time.
  • Run prior-art and FTO searches at ideation, not at filing. Discovering a blocking patent at the disclosure-review stage is expensive. Discovering it before engineering begins is free. Modern AI-driven semantic search has collapsed the time and cost of running these searches, which removes the historical excuse for doing them late.
  • Use defensive publishing as a deliberate portfolio tool. Not every invention should be a patent application. Establishing prior art on incremental features can preserve your freedom to operate, deny competitors the ability to fence you in, and free patent budget for the inventions that actually deserve protection.

This is the work IP.com was built for. Our InnovationQ+ platform uses our proprietary Semantic Gist® engine to run natural-language landscape and whitespace searches across patent and non-patent literature, including exclusive IEEE technical content, giving R&D and IP teams a clear picture of where invention opportunity exists before ideation begins. From there, IQ Ideas+ with CompassAI takes teams through structured ideation with prior art comparison and claims mapping built in, so you’re not designing into a thicket you haven’t seen yet. Our Patent Landscape, Freedom to Operate, and Portfolio Intelligence services give IP and R&D leaders the expert analysis and structured findings they need to make strategic calls about where to invent, where to design around, and where to publish defensively. And the Prior Art Database lets innovators establish defensive disclosures in hours, not months.

The companies that will win the next decade aren’t the ones with the biggest patent portfolios. They are the ones who know, earlier and more precisely than their competitors, exactly where the brambles are and how to route around them. The thicket isn’t going away. The good news is, neither is the map.

Sources

Share This

Book a Custom Defensive Publishing Demo

Book a Custom Innovation Q+ Demo

Book a Custom IQ Ideas+ Demo

Book a Custom Demo