KSR v. Teleflex and the Art of Cross-Domain Invalidity

The Problem With How Most Invalidity Searches Are Run

Most invalidity searches begin in the wrong place: searching the same technical domain as the patent being challenged, using the same terminology the patent drafter used, in the same databases that anyone would think to search. This approach has a structural flaw — if the prior art lived in the obvious places, using the obvious search terms, in the most obvious databases, there is a reasonable chance the patent examiner already found it during prosecution.

The prior art that defeats patents — the art that either anticipates under 35 U.S.C. § 102 or renders obvious under § 103 — often exists precisely where the patent drafter did not look. It exists in adjacent technical domains. It exists in industry-specific literature that patent examiners rarely read. It exists in standards documents, conference proceedings, doctoral dissertations, and expired patents in domains that seem entirely unrelated to the claimed invention.

Understanding why this is true, and how to exploit it systematically, requires understanding what the Supreme Court actually said in KSR International Co. v. Teleflex Inc. (2007) — and understanding it at a technical level, not just a legal one.

KSR v. Teleflex: The Technical Context

The patent in dispute — Teleflex's U.S. Patent 6,237,565 — claimed an adjustable car pedal assembly with an electronic sensor positioned on the fixed pivot point of the pedal to detect pedal position. This sounds specific, but the claim was essentially: take an adjustable pedal (prior art — Asano 1996), add an electronic sensor (prior art — Rixon 1994), and locate the sensor on the fixed pivot rather than on the movable assembly (a design choice).

KSR argued the combination was obvious. Teleflex argued that under the then-prevailing rigid Teaching-Suggestion-Motivation (TSM) test, you needed explicit prior art evidence that a person skilled in the art would have been motivated to combine Asano's adjustable pedal with Rixon's fixed-pivot sensor. The Federal Circuit agreed with Teleflex. The Supreme Court reversed unanimously.

What KSR Actually Said — The Flexible Approach

The Supreme Court's KSR opinion articulated several principles that fundamentally changed how § 103 obviousness works in practice:

• A person of ordinary skill in the art is not a robot — they have creativity and common sense. They combine prior art references because it makes engineering sense to do so, not because a document explicitly tells them to combine them.
• Obvious to try is sometimes obvious — when there are a finite number of identified, predictable solutions to a known problem, choosing among them is obvious even without a specific teaching to do so.
• Known problem + known solution in the prior art = obvious combination — if the problem being solved by the patent was known and a solution existed in a different technical domain, combining them is likely obvious even without domain-specific cross-reference.
• Market forces drive obvious combinations — if designers in two different industries were simultaneously converging on the same solution, the combination is obvious regardless of whether anyone wrote a paper explicitly suggesting it.

The Court of Appeals erred in concluding that courts and patent examiners should look only to the problem the patentee was trying to solve... [A] patent composed of several elements is not proved obvious merely by demonstrating that each of its elements was, independently, known in the prior art. But it can be shown to be obvious by demonstrating that there existed a reason to combine the known elements in the fashion claimed by the patent at issue. — Justice Kennedy, KSR v. Teleflex (2007)

The Cross-Domain Implication: Where the Prior Art Actually Lives

KSR's most powerful practical implication is that prior art from a completely different technical domain is valid § 103 prior art if a person of ordinary skill would have been motivated to look there — and the bar for "motivation" under the post-KSR flexible approach is substantially lower than under rigid TSM.

Consider a patent claiming a wireless sensor node with an ultra-low-power wake-up receiver that monitors a specific frequency band and triggers the main radio only when a valid packet is detected. The patent is filed by a smart home company, claiming the domain of home automation IoT. Where does the prior art actually live?

• Military and defence literature — Wake-up receivers for military sensor network nodes were extensively developed in DARPA-funded research from 2001-2008 (SensorNet, WINS programs). Papers from IEEE ISSCC, CICC, and RFIC symposia covered sub-μW wake-up receiver design long before smart home IoT became an industry.
• Medical implant telemetry — Implantable medical devices have used ultra-low-power wake-up receivers since the 1990s to enable communication-on-demand without continuously active radios draining the non-rechargeable battery. The constraint (minimise quiescent current of the receive chain) is identical to IoT; the implementation approaches are prior art.
• RFID and contactless smart card standards — ISO 14443 contactless smart card readers implement a form of passive wake-up at the physical layer; backscatter communication in UHF RFID (EPC Gen2) uses similar RF envelope detection principles.

A searcher who looked only at IoT patents and papers would find nothing useful. A cross-domain searcher applying KSR methodology finds multiple independently valid combinations that render the claims obvious.

How We Apply KSR Methodology at Bullseye

Every invalidity search we execute begins with the same structured analysis before any database search begins:

• Step 1: Claim decomposition by function — break every independent claim into its functional elements, stripping terminology to its underlying engineering function.
• Step 2: Domain mapping — for each functional element, map the technical domains where that function has been implemented, regardless of industry. The "domains" are almost always broader than the patent's nominal field.
• Step 3: KSR motivation analysis — for each domain, assess the motivation to combine: Was the problem the patent is solving known in that domain? Was the solution known? Would a practitioner have known to look there?
• Step 4: Database strategy — only after steps 1-3 is any database searched. IEEE Xplore, ACM Digital Library, SpringerLink, PubMed, arXiv, 3GPP document servers, ETSI, NASA Technical Reports Server, DTIC (Defence Technical Information Center) — all are searched as standard practice.
• Step 5: Claim mapping for combined-art § 103 grounds — mapping the combined art to the claim elements with explicit KSR motivation language supporting the combination.

A Real Example: Semiconductor Packaging and Food Science

In one landmark invalidity case (details anonymised), a patent claiming an underfill adhesive composition for flip-chip IC packaging — specifically a mixture of epoxy resin with specific particle size distribution of silica filler and a reactive diluent — was found to have § 103 prior art in food science literature. The specific combination of particle packing efficiency algorithms used in the semiconductor packaging composition had been published in the Journal of Food Engineering as a method for optimising flour particle distributions in bread baking. The POSITA in semiconductor packaging would not routinely read food science journals — but a KSR-flexible court found the motivation existed: the engineering principle (maximising packing efficiency of particles in a matrix) is field-agnostic, and a material scientist formulating underfill compounds would have consulted colloidal science and particle packing literature that spans food science, cosmetics, and pharmaceuticals.

Practical Implications for Patent Attorneys

For patent attorneys commissioning invalidity searches, the KSR framework has direct practical implications for how search scope should be defined:

• Do not limit search scope to the patent's stated field of use — that is where the patent drafter looked, and where prior art is most likely to have been found during examination.
• Request explicit cross-domain mapping as part of every invalidity search deliverable — which other technical domains implement the same functional elements?
• Ask the search firm to explicitly identify § 103 combination grounds and articulate the KSR-flexible motivation for each combination — not just identify individual references.
• For technical patents in hardware domains (semiconductor, wireless, power electronics), always include IEEE and non-patent literature — the most critical prior art in engineering domains often predates patent filing in academic or standards literature.

"The prior art that defeats patents most effectively almost always comes from a technical domain the patent drafter never considered. That is not a coincidence — it is a structural consequence of how patents are examined and where examiners search. Our job is to look precisely where the examiner did not." — Udeyanju Kumar, Bullseye