COLUMN // PATENT
When to Disclose an Invention
Patent law makes the U.S. a first-to-file system, but exceptions abound.
Written by Kirk Teska
Patent law is established to protect intellectual property, and it works best when the rules are roughly the same everywhere. In 2011, as part of a harmonization effort, the United States changed from a first-to-invent to a first-to-file (a patent application) regime primarily because that’s the system of almost every other country.
The first-to-invent system in place for many years did produce some fights among competing inventors for a given patent—fights governed by a set of byzantine rules—but usually the first patent application filer prevailed. That meant the U.S. was a de facto first-to-file system anyway especially since these fights were few and far between. Think about it: What are the odds two inventors invent almost exactly the same thing at the same time?
The American Invents Act that implemented the first-to-file system solved a problem that didn’t really exist.
The weird part of the regime change was the U.S. kept its “one year grace period” system which almost no other country has. So much for harmonization. In the U.S., an inventor can invent something, publicly disclose it, sell it, whatever, and still file a valid patent application up to a year later. But, to make the one-year grace period system work in the new first-to-file system, you have to allow a first-in-time inventor but second-in-time patent application filer to prevail against a first-in-time patent application filer in certain scenarios. So, the patent laws allow the second-in-time filer, if he publicly discloses the invention before the first-in-time filer files his patent application, to not have that patent application knock out the first inventor’s second-in-time patent application.
For any sadomasochists out there who want to dive deeper into this, it’s all in 35 United States Code section 102. Beware: there are five subsections, 11 sub-subsections, and even a few sub-sub-subsections. If there is a more poorly written law in our books, I’ve not seen it. My patent law students seem to like my crude but easy to memorize summary: Absent theft or some bizarre fact pattern, the first patent application filer wins unless the first inventor, second patent application filer discloses before the second inventor files (and then also files within a year after the disclosure). These two charts may help.
Scenario 1 represents the first-to-file regime where the first filed patent application wins even though that invention was second in time over the first inventor’s invention. An exemption to that rule is shown in the next scenario.
Scenario 2 shows how, under the one-year grace period system, since the first inventor disclosed the invention before the second inventor filed a patent application, the second inventor’s first-to-file patent application does not count against the first inventor’s second-in-time patent application. Here, in contrast to scenario 1, the second filer prevails. The second scenario involves a “protective disclosure” by the first inventor who disclosed the invention before the other inventor filed a patent application. The earlier disclosure protects the second filed patent application against the first filed patent application so long as the discloser complies with the one-year grace period and follows up with a patent application within a year of the disclosure.
Let’s now look at one more scenario:
In this scenario, the one-year grace period does not help the inventor because the inventor disclosed the invention and more than one year later filed a patent application that will be held invalid. Thus, scenario 3 involves an “invalidating disclosure.”
So far, so good except for one little problem: Recently, the Patent Court ruled protective disclosures in scenario 2 do not mean the same thing as invalidating disclosures in scenario 3. For example, a private (“secret”) sale of a product incorporating an invention would count as an invalidating disclosure under scenario 3 but not as protective disclosure under scenario 2. The same would probably be true of a secret process used to manufacture a product that was then sold—which would count as an invalidating disclosure (as to both the process and the product) but probably not a protective disclosure as to the process.
The reason is invalidating disclosures and protective disclosures involve different policy considerations.
The policy under scenario 2 is to protect an inventor who publicly discloses an invention from patent application filers who beat the inventor in the race to the Patent Office. The policy under scenario 3 is to prevent an inventor from commercializing an invention and then waiting too long to file a patent application even when, in the commercialization effort, no one learns about the nature or details of the invention.
Does it matter? Maybe, because certain patent strategies could be adversely affected. IBM and others, for example, engage in invention pledge programs where they publish and disclose technologies they decide not to patent to protect themselves from later patent application filers attempting to protect the same technology. Also, some companies publish first and then wait a year to file a patent application to take full advantage of the one-year grace period (enabling further research and development and making sure the invention really has market value) to protect themselves from others who file a patent application before they do.
Given the Court’s recent ruling, some of these strategies may need revamping.
“Absent theft or some bizarre fact pattern, the first patent application filer wins unless the first inventor, second patent application filer discloses before the second inventor files (and then also files within a year after the disclosure).”
For an invalidating disclosure, the actual invention might never be known to others (such as the sale of a product embodying invention even if the sale is private), public use of the invention (even though the public never sees its details), and even an offer for sale of a product embodying the invention. For a valid protective public disclosure, the publication, sale, or use must involve the public’s actual knowledge of the details of the invention. Thus, a scientific article describing the invention works for both an invalidating disclosure and a protective disclosure while showing a gizmo (but not its internal workings) at a trade show constitutes an invalidating disclosure but not a protective disclosure.
Until this case was decided, many people thought that what works for an invalidating disclosure also works for a protective disclosure. That’s not true anymore. Maybe a patent strategy that predates even the change to the first-to-file regime may again be invoked: file early and often.
KIRK TESKA is the managing partner of Iandiorio, Teska and Coleman, LLP, an adjunct professor at Suffolk Law, and the author of two books: Patent Savvy for Managers (NOLO) and Patent Project Management (ASME Press).
© 2025 The American Society of Mechanical Engineers. All rights reserved.