12 Landmark Patent Infringement Cases & Analysis

Patent law shapes how companies protect their innovations and defend against competitors, but the rules governing patent disputes didn’t emerge from a vacuum. They were built case by case, through landmark patent infringement cases that redefined what can be patented, how damages are calculated, and what remedies are available when someone crosses the line.

In this guide, I’ll walk through 12 famous patent infringement cases, from genetically engineered bacteria to smartphone design wars, and evaluate the impact each has had on patent law, enforcement, and strategy.

Key Takeaways from 12 Famous Patent Infringement Cases

1. Patent infringement law is built on a handful of foundational Supreme Court decisions. 

Patent infringement court cases like Diamond v. Chakrabarty, eBay v. MercExchange, and Microsoft v. i4i established the baseline rules for what can be patented, what happens when you win, and how hard it is to invalidate a patent. Understanding these decisions is essential for anyone involved in patent litigation or IP strategy.

2. Damages and remedies are among the most actively evolving areas of patent law. 

From Carnegie Mellon’s billion-dollar royalty verdict to Samsung v. Apple’s redefinition of design-patent profits, courts continue to refine how infringement damages are measured. And the stakes are enormous.

3. Modern patent disputes increasingly involve overlapping IP theories and cross-border complexity. 

Cases like Google v. Oracle and Microsoft v. AT&T show that infringement litigation now spans multiple areas of intellectual property and raises questions about territorial reach that didn’t exist a generation ago.

Managing the complexity of patent litigation starts with the right technology. Learn how DISCO’s AI-powered ediscovery platform helps legal teams navigate massive document volumes, prior art review, and expert preparation in large patent cases. 

12 Landmark Patent Infringement Cases and Analysis

1. Diamond v. Chakrabarty (1980)

In the late 1970s, microbiologist Ananda Chakrabarty developed a genetically engineered bacterium capable of breaking down multiple components of crude oil, something no naturally occurring organism could do. 

When he filed for a patent on the bacterium itself, the U.S. Patent Office rejected the application, arguing that living organisms fell outside the scope of patentable subject matter under 35 U.S.C. § 101.

The case reached the Supreme Court, which faced a deceptively simple question: Can a living, human-made organism be patented? In a 5-4 decision, the Court said yes. Chakrabarty’s bacterium qualified as a “manufacture” or “composition of matter” because it had markedly different characteristics from anything found in nature. 

The ruling opened the door for patent protection across biotechnology, laying the groundwork for decades of innovation in genetic engineering, pharmaceuticals, and synthetic biology.

Key points

• Living organisms can be patented if they are human-made and have characteristics distinct from anything found in nature.
• The decision is widely credited with enabling the modern biotechnology industry.
• Patent law focuses on what people invent, not whether the invention is mechanical, chemical, or biological.

Case analysis

Chakrabarty is the foundational case on patentable subject matter for biotechnology. The Court interpreted § 101’s categories broadly, citing congressional intent to encourage innovation and the famous “anything under the sun that is made by man” language from the legislative history. It drew a clear line between unpatentable natural phenomena and human-made inventions exhibiting “markedly different characteristics” from what exists in nature.

While later decisions — most notably Mayo Collaborative Services v. Prometheus Laboratories and Association for Molecular Pathology v. Myriad Genetics — narrowed eligibility for natural correlations and isolated DNA, Chakrabarty remains the key authority supporting patent eligibility for genetically modified organisms and engineered biological constructs. 

In patent infringement litigation, the case frequently surfaces in validity challenges when accused products involve recombinant organisms, engineered cells, or synthetic biology tools. It anchors the argument that such constructs remain within the patent-eligible domain even as other subject-matter doctrines have tightened.

2. eBay Inc. v. MercExchange, L.L.C. (2006)

MercExchange held a patent on online auction technology and sued eBay for infringement. A jury agreed and awarded damages, but then the case took an unexpected turn. 

The district court refused to issue a permanent injunction, noting that MercExchange didn’t practice the patent itself and had shown a willingness to license it. The Federal Circuit reversed, applying its longstanding rule that winning patentees are generally entitled to injunctions.

The Supreme Court stepped in and unanimously vacated the Federal Circuit’s decision. The Court held that patent cases are not exempt from the traditional four-factor test for permanent injunctions: the plaintiff must show irreparable injury, inadequacy of monetary damages, a balance of hardships favoring an injunction, and that the injunction serves the public interest. No categorical rules, whether for or against any class of patentee, apply in this case.

Key points

• Winning a patent infringement lawsuit does not automatically entitle the patent holder to a court order blocking the infringer.
• Courts must weigh fairness and practical impact before granting an injunction.
• The decision made injunctions significantly harder to obtain, especially for non-practicing entities.

Case analysis

eBay dismantled the Federal Circuit’s near-automatic injunction rule for prevailing patentees and replaced it with the same equitable framework used across other areas of law. 

The practical effect has been substantial: post-eBay, courts routinely deny injunctions to patent holders who primarily license rather than practice their patents, and even practicing entities face meaningful hurdles when the patented feature is a small component of a complex product.

For litigators, the decision elevated the importance of early, evidence-backed irreparable-harm narratives and careful framing of competitive dynamics. It also established ongoing royalties as a viable alternative remedy, creating a new category of post-verdict damages disputes. 

For accused infringers, eBay provides critical leverage in settlement negotiations. If the threat of an injunction is uncertain, the patentee’s bargaining position weakens. The case remains one of the most frequently cited patent decisions in federal courts.

3. Bowman v. Monsanto Co. (2013)

Indiana farmer Vernon Bowman bought commodity soybeans from a local grain elevator, beans that largely carried Monsanto’s patented Roundup Ready trait, and replanted them without a license. His reasoning was straightforward: Monsanto’s patent rights should have been “exhausted” after the first authorized sale, giving him the right to use the seeds however he wished, including replanting.

The Supreme Court unanimously disagreed. Patent exhaustion, the Court held, applies to a particular item that has been sold, not to copies or reproductions of the patented invention. 

By planting and harvesting the seeds, Bowman was making new copies of Monsanto’s patented technology, which is fundamentally different from simply using a purchased product. The ruling reinforced patent protection for self-replicating technologies, with implications well beyond agriculture.

Key points

• Buying a patented product doesn’t grant permission to reproduce it, even when the product naturally self-replicates.
• Patent rights can extend to self-replicating technologies like seeds, biological materials, and potentially software.
• Farmers and businesses must respect license terms, even when dealing with commodity products.

Case analysis

Bowman clarified a critical boundary of the patent exhaustion doctrine: exhaustion allows a purchaser to use or resell a patented article but does not permit deliberate reproduction. 

The Court’s reasoning distinguished between using a purchased item (protected by exhaustion) and making new copies (an act of infringement), and while the opinion is formally limited to Bowman’s seed-replanting strategy, it sends a clear signal about any technology where replication is the primary means of obtaining value.

For patent litigators, the case reinforces the viability of infringement claims against downstream actors who exploit self-replicating products beyond the scope of their licenses. It also informs transactional strategies for drafting seed-use agreements, biologics licenses, and software distribution terms that clearly restrict unauthorized reproduction. 

The opinion leaves some room for future disputes over inadvertent or incidental replication, but its core holding strongly favors patentees in agricultural biotech and related fields.

4. Microsoft Corp. v. i4i Limited Partnership (2011)

i4i, a small Canadian technology company, sued Microsoft for patent infringement, alleging that a custom XML editing feature in Microsoft Word violated i4i’s patent. A Texas jury agreed, awarding substantial damages and triggering an injunction that briefly barred sales of certain Word versions in the United States.

Microsoft appealed, challenging not just the verdict but the standard of proof itself. It argued that when prior art wasn’t considered by the Patent Office during examination, accused infringers should only need to prove invalidity by a preponderance of the evidence — a lower bar than the clear-and-convincing standard traditionally applied. 

The Supreme Court rejected that argument and affirmed the clear-and-convincing standard across the board, holding that the presumption of validity codified in § 282 demands it, regardless of whether the examiner reviewed the prior art at issue.

Key points

• Patents are presumed valid, and challenging that validity in court requires clear and convincing evidence.
• Even major software companies can be ordered to change or stop selling key products.
• Small innovators can successfully enforce patents against industry giants.

Case analysis

Microsoft v. i4i settled a long-running debate about whether the burden of proof for patent invalidity should shift depending on what the Patent Office actually reviewed. By affirming the uniform clear-and-convincing standard, the Court entrenched the difficulty of invalidating patents at trial, even when new prior art surfaces that the examiner never saw. 

The Court acknowledged, however, that such art may carry significant weight within that framework, giving juries latitude to assess the evidence accordingly.

For practitioners, i4i heightens the strategic importance of inter partes review and other PTAB proceedings as parallel forums where the burden of proof is lower (preponderance of the evidence). 

The decision is frequently cited in jury-instruction disputes and appellate briefing about whether district courts properly framed the invalidity standard, and it remains a key consideration in every patent case where validity is contested.

5. Carnegie Mellon University v. Marvell Technology Group (2012)

Carnegie Mellon University sued chipmaker Marvell Technology, alleging that Marvell’s read-channel chips for hard disk drives infringed two CMU patents on methods for improving the accuracy of reading data from magnetic media. 

What began as a university enforcing its research patents became one of the largest patent verdicts in U.S. history: a Pittsburgh jury awarded approximately $1.17 billion in reasonable-royalty damages.

The district court enhanced the damages based on willfulness. Marvell had known about and consciously used the patented technology for nearly a decade without obtaining a license. The case ultimately settled for $750 million while appeals were pending. 

Beyond the headline number, the case raised important questions about how royalties should be calculated when a small per-unit rate is applied across billions of chips, including those sold outside the United States.

Key points

• University patents can lead to massive payouts when industry uses the technology without a license.
• Chip-level innovations embedded deep inside products can be just as valuable as visible consumer features.
• Willful copying over extended periods dramatically increases a company’s financial exposure.

Case analysis

Carnegie Mellon is a landmark damages case on reasonable-royalty calculation and territorial scope. The jury’s 50-cents-per-chip royalty, applied to billions of units, raised hard questions about apportionment, comparable licenses, and the role of foreign sales in U.S. patent damages. 

On appeal, the Federal Circuit largely upheld the infringement and validity findings but refined the damages framework, emphasizing the need to tie royalty bases and rates to the value of the patented contribution, not the entire product’s revenue.

The court also addressed laches, ultimately allowing substantial back-damages despite CMU’s delay in filing suit because Marvell had consciously copied the technology for years. For practitioners, the case underscores the importance of robust damages evidence, careful selection of royalty base and rate, and the strategic leverage that universities and research institutions can wield when their core research underpins widespread commercial products.

Cases like Carnegie Mellon involve massive volumes of technical documentation, prior art, and communications spanning years of development. DISCO’s cloud-native platform processes terabytes of data at enterprise scale, and tools like Cecilia Q&A let legal teams ask targeted questions across their entire document set, surfacing critical evidence in seconds instead of weeks. Learn more about DISCO Ediscovery →

6. Apple Inc. v. Samsung Electronics / Samsung Electronics Co. v. Apple Inc. (2012/2016)

Few patent disputes have captured public attention like the Apple-Samsung smartphone wars. In 2011, Apple sued Samsung in federal court, alleging that Samsung’s Galaxy phones and tablets copied the iPhone’s design, from the rounded-corner front face and home button layout to software features like bounce-back scrolling and tap-to-zoom. 

Apple asserted both design and utility patents as well as trade dress claims, building a case that spanned the full range of IP protection. A California jury largely agreed, finding willful infringement and awarding over $1 billion in damages, later adjusted on remittitur and retrial. 

Samsung appealed the design-patent damages — about $399 million — to the Supreme Court, arguing that it shouldn’t owe its entire profit on infringing smartphones when the patented designs covered only specific visual elements, not the whole device.

The Supreme Court unanimously agreed with Samsung on the damages question, holding that the “article of manufacture” for purposes of calculating design patent profits under 35 U.S.C. § 289 can be a component of a product, not necessarily the product sold to consumers. 

The Court vacated the Federal Circuit’s rigid approach and remanded for further proceedings, though it declined to specify a definitive test for identifying the relevant article.

Key points

• Design details like icons, screen shapes, and interface effects can be protected by patent — not just overall design of the full finished product.
• After a big jury verdict, appeals can dramatically change how damages are calculated.
• Design patents don’t necessarily entitle owners to an infringer’s entire profit on a finished product.

Case analysis

Taken together, these cases illustrate both the offensive power and the limits of overlapping IP strategies. At trial, Apple’s success demonstrated how carefully curated damages narratives, jury-friendly visuals, and evidence of intent can drive findings of willful infringement and massive monetary relief. The Federal Circuit’s treatment of claim scope, willfulness, and functionality limits on trade dress shaped the playbook for smartphone litigation, and more broadly, consumer-tech IP strategy.

At the Supreme Court, Samsung v. Apple became the leading modern authority on design patent damages. The Court rejected the view that the “article of manufacture” must always be the entire product sold to consumers and adopted a two-step framework: 

1. Identify the relevant article of manufacture;
2. Calculate the infringer’s total profit on that article. 

Post-Samsung, lower courts and practitioners have developed multifactor approaches focusing on physical separability, prominence of the design, and conceptual distinctness between the patented design and the rest of the product.

For litigators, the combined decisions refocused damages strategy on granular product teardown, detailed accounting testimony, and careful claim drafting around discrete visual elements that can plausibly stand alone as articles of manufacture.

7. Amgen Inc. v. Sanofi (2023)

Amgen and Sanofi clashed over patents covering PCSK9-inhibitor antibodies, a class of biologic drugs used to lower LDL cholesterol. Amgen held broad “genus” patents that claimed any antibody binding certain residues on the PCSK9 protein and blocking its interaction with LDL receptors. 

Under that definition, the claims potentially covered millions of distinct antibodies, though the patents described only a limited number of specific examples and a trial-and-error screening process for finding others.

Sanofi argued the patents were invalid for lack of enablement, and the Supreme Court unanimously agreed. The claims failed the enablement requirement under 35 U.S.C. § 112 because the patents didn’t teach a skilled practitioner how to make and use the full scope of what was claimed — only a subset, plus broad functional guidance that amounted to little more than a research roadmap.

Key points

• Drug patents can try to cover entire families of molecules, but courts require the patent to teach others how to make what it claims.
• Even blockbuster biologic patents can be struck down if they are too broad relative to what the specification actually describes.
• The decision tightened the enablement requirement, particularly for functionally defined claims in the life sciences.

Case analysis

Amgen reaffirmed that § 112 demands that a patent enable a person of ordinary skill in the art to make and use the full scope of the claimed invention without undue experimentation, not just a subset plus functional guidance. 

The Court framed the holding as a straightforward application of longstanding enablement principles, but its practical impact on biotech and pharmaceutical patent strategy is significant. Broad functional claims, especially those defined by what an antibody does rather than what it is, are now substantially more vulnerable at the invalidity stage.

For accused infringers, Amgen strengthens leverage in both litigation and settlement whenever asserted claims resemble the functionally drafted genus claims at issue here. 

For patent holders, the decision encourages more modest claim drafting tethered to structurally characterized species or well-defined sub-genera, and it makes thorough specification detail more critical than ever. 

The case will likely reshape how pharmaceutical and biotech companies approach both prosecution and enforcement for years to come.

8. Google LLC v. Oracle America, Inc. (2021)

Oracle sued Google over Google’s use of Java APIs in the Android operating system. The dispute began with both copyright and patent infringement claims, though the patent issues eventually fell away, leaving a high-stakes fight over whether Google’s reimplementation of roughly 11,500 lines of Java API declaring code was protected by fair use.

After a long procedural history that included multiple trips to the Federal Circuit, the Supreme Court assumed, without deciding, that Oracle’s copyrights were valid and held that Google’s use constituted fair use as a matter of law. 

The Court emphasized the functional nature of the API code, Google’s transformative purpose in building a new smartphone platform, and the risk of locking in programmers who had invested years learning the Java language.

Key points

• Software disputes often involve overlapping patent and copyright claims, especially around programming interfaces.
• Reusing functional programming interfaces to build new platforms may qualify as fair use.
• Multi-year IP battles can shape how entire technology ecosystems develop and compete.

Case analysis

Though primarily a copyright decision, Google v. Oracle provides important context for software-related patent disputes. The case highlights a fundamental tension in intellectual property law: functionality cuts in opposite directions depending on the type of IP at issue. 

In copyright (and trade dress), functional elements are typically unprotectable. And if there are only a few ways to express an idea, the expression is considered a function of the idea itself. Utility is a limiting principle. 

In patent law, functionality is the entire point. 

This dichotomy means that for complex software products, the same feature might be analyzed under entirely different legal frameworks depending on whether the claim sounds in patent or copyright.

For patent litigators working in technology, the decision underscores the strategic value of alternative or overlapping IP theories and the importance of understanding how courts view functional elements across different IP regimes. 

While Google doesn’t limit software patents directly, it operates in the same conceptual space as Alice-era subject-matter scrutiny, pushing parties to think carefully about which form of IP protection best fits their technology and where the boundaries lie.

9. Microsoft Corp. v. AT&T Corp. (2007)

AT&T held a patent on speech-codec technology and alleged that Microsoft’s Windows operating system infringed it. The central question wasn’t whether Windows used the patented technology — it did — but whether Microsoft could be held liable for copies of Windows installed on computers manufactured and sold overseas.

The case turned on 35 U.S.C. § 271(f), which imposes infringement liability for supplying “components” of a patented invention from the United States for assembly abroad. 

AT&T argued that the master Windows disk sent overseas was such a component. The Supreme Court disagreed, holding that software in the abstract, or a master disk sent abroad and replicated there, is not a “component” supplied from the U.S. under § 271(f). The physical copies came into existence overseas. They were not exported from the United States.

Key points

• U.S. patent law does not automatically cover foreign sales of products incorporating U.S.-developed software.
• Shipping code overseas for local duplication is treated differently than exporting finished hardware.
• Global technology companies must carefully structure manufacturing and licensing to manage cross-border patent risk.

Case analysis

Microsoft v. AT&T constrains the extraterritorial reach of U.S. patent law under § 271(f). The Court rejected the Federal Circuit’s broader view that intangible software code qualifies as a supplied “component” when transmitted abroad and copied onto foreign hardware, emphasizing the presumption against extraterritorial application of U.S. statutes. 

Only the physical embodiment installed on a machine — not the abstract code — could qualify, and that embodiment was created overseas.

For litigators, the decision limits damages theories that seek to capture worldwide sales based on domestic software authoring or master-disk export. It encourages granular analysis of where accused components are manufactured and combined, and it remains a key citation in cross-border patent infringement disputes. 

For companies, the case informs how international distribution, OEM relationships, and code deployment should be structured to minimize exposure under § 271(f).

10. Minerva Surgical, Inc. v. Hologic, Inc. (2021)

Minerva and Hologic disputed patents covering devices for endometrial ablation, a medical procedure used to treat abnormal uterine bleeding. 

The key issue before the Supreme Court was not the infringement itself but assignor estoppel, an equitable doctrine preventing an inventor who assigns a patent to someone else from later arguing the patent is invalid when they’re sued for infringement.

The Court declined to abolish the doctrine but narrowed it. Assignor estoppel applies only when the assignor’s later validity challenge is inconsistent with representations made at the time of assignment. If the claims were materially broadened after the assignment, or if the assignment occurred before specific claims existed (such as blanket assignments of future inventions), estoppel may not apply.

Key points

• Inventors who assign their patents can’t always turn around and argue those patents are invalid.
• The doctrine has limits: it doesn’t apply when claims are materially changed after the assignment.
• Medical-device disputes often raise both technical and fairness questions about who can challenge a patent.

Case analysis

Minerva reshaped assignor estoppel without eliminating it. The Court grounded the doctrine in principles of fair dealing, holding that it should apply when an assignor makes an implicit or explicit representation of validity and then takes an inconsistent position in subsequent litigation. 

But the decision identified important carve-outs: estoppel may not apply when claims are materially broadened after assignment, when the assignment covers future inventions not yet claimed, or when the PTO itself has confirmed a claim’s validity in a way that changes the equitable calculus.

In practice, Minerva has focused litigants on whether post-assignment claim amendments create enough divergence to preserve invalidity defenses. It encourages careful drafting of assignment agreements, strategic use of continuation practice, and early analysis of who might be estopped from challenging validity in future patent infringement suits involving the same patent family.

11. Akamai Technologies, Inc. v. Limelight Networks, Inc. (2014/2015)

Akamai held a patent on a method for delivering content over the internet using a content delivery network (CDN). Limelight operated a competing CDN and performed most, but not all, of the patented method’s steps. 

The final step, known as “tagging,” was carried out by Limelight’s customers rather than Limelight itself. Akamai argued that Limelight should be liable for inducing its customers to perform that remaining step, even though no single party performed every step of the claimed method.

The Supreme Court unanimously rejected that theory. Inducement liability under § 271(b) requires that someone actually committed direct infringement under § 271(a), and if no single entity performed all the steps, there was no direct infringement to induce. 

On remand, however, the Federal Circuit expanded the circumstances under which one party’s actions can be attributed to another for direct-infringement purposes, ultimately reinstating the jury’s $40 million-plus verdict against Limelight.

Key points

• There can be no induced infringement without underlying direct infringement by someone.
• Companies cannot evade patent liability simply by having customers perform certain steps of a patented method.
• Method patent claims should be drafted carefully to avoid divided-infringement vulnerabilities.

Case analysis

The Akamai litigation, which spanned nearly a decade and multiple trips between the Federal Circuit and the Supreme Court, reshaped the law of divided infringement. 

The Supreme Court’s 2014 decision established a firm structural principle: inducement liability under § 271(b) requires direct infringement under § 271(a). Without a single entity committing direct infringement, there is nothing to induce.

But the story didn’t end there. On remand, the Federal Circuit expanded the scope of direct infringement by holding that one actor can be held responsible for another’s performance of method steps when the alleged infringer “conditions participation in an activity or receipt of a benefit upon performance of a step or steps of a patented method, and establishes the manner or timing of that performance.” 

This broadened attribution doctrine gave the Federal Circuit a way to close the divided-infringement loophole without overriding the Supreme Court’s holding on inducement.

For practitioners, the case underscores the importance of claim drafting: method patents should, whenever possible, be structured so that all steps are performed by a single entity. It also informs litigation strategy around attribution theories and contract-based arguments for imputing one party’s actions to another.

Complex patent cases like these generate enormous volumes of evidence — from prior art and prosecution histories to engineering files and cross-border communications. CO Auto Review uses generative AI to perform first-pass document review at scale, with natural-language justifications for every tagging decision. Combined with DISCO’s high-speed search and AI-prioritized review, legal teams can move from data to strategy faster and more efficiently. Explore Cecilia AI →

Why these decisions still shape every patent dispute

These 12 cases represent more than legal history. They are the operating rules of modern patent litigation, covering patent law issues ranging from what qualifies as a patentable invention to how damages are measured, from the limits of injunctive relief to the boundaries of induced and divided infringement. These decisions define the framework within which every patent dispute plays out.

What connects them is a common thread: patent law is constantly adapting to new technologies, new business models, and new strategies for both asserting and defending against infringement. 

The courts that decided these cases were grappling with genetically modified organisms, self-replicating seeds, software interfaces, smartphone designs, and internet delivery networks, and the doctrines they established continue to evolve as new technologies raise new questions.

For legal teams navigating this complexity, the right tools matter as much as the right strategy. DISCO’s AI-powered ediscovery platform excels under the demands of patent litigation, effortlessly processing massive volumes of technical data and accelerating document review with generative AI.

Ready to see how DISCO can transform your patent litigation workflow? Request a demo →