For a university technology transfer office (TTO), 35 U.S.C. § 102 is not an abstract statute — it is the clock that determines whether a faculty inventor’s disclosure, conference talk, or preprint has already destroyed the patent rights the office is trying to protect. This guide explains what § 102 actually requires, how the America Invents Act’s one-year grace period works (and where it doesn’t reach), and how those deadlines interact with the internal invention-disclosure and Bayh-Dole election-of-title timelines every federally funded university lab runs on.
What 35 U.S.C. § 102 covers
35 U.S.C. § 102, titled “Conditions for patentability; novelty,” is the statutory basis for one of the two core patentability requirements (the other being non-obviousness under 35 U.S.C. § 103). In plain terms: an invention is not patentable if it was already disclosed to the public, by anyone, before the applicant’s effective filing date — subject to a narrow set of exceptions for the inventor’s own recent disclosures.
The current text of § 102 is the version created by the Leahy-Smith America Invents Act (AIA), effective for any patent application with an effective filing date on or after March 16, 2013. The AIA replaced the pre-2013 “first-to-invent” system with a first-inventor-to-file system. This is a distinction research administrators need to hold onto: priority is no longer about who conceived the idea first in a lab notebook — it is about who has the earliest effective filing date, with a specific carve-out (below) for the inventor’s own disclosures.
What counts as prior art under § 102(a)
Section 102(a) defines two categories of prior art that can defeat novelty:
- § 102(a)(1) — the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public, before the effective filing date. Note this category is not limited to disclosures by others: the inventor’s own paper, poster, conference talk, thesis defense, or sale can count against them under (a)(1) unless a (b)(1) exception applies (see below).
- § 102(a)(2) — the claimed invention was already described in a U.S. patent, U.S. patent application publication, or PCT application designating the U.S., that names another inventor and has an earlier effective filing date. This is the “someone else beat you to the patent office” category, sometimes shortened to “secret prior art” because the competing application may not have published yet when your client files.
“Effective filing date” is a term of art: it is the earliest of (1) the actual filing date of the application, or (2) the filing date of the earliest application to which priority or benefit is properly claimed (a provisional application, a foreign priority application, or a parent non-provisional). This is precisely why filing a provisional patent application promptly — even a bare-bones one — is one of the few tools a TTO has for locking in an earlier effective filing date once a disclosure is imminent or has already happened.
The one-year grace period under § 102(b) — and its real limits
§ 102(b)(1) provides the AIA’s inventor grace period: a disclosure that would otherwise count as § 102(a)(1) prior art is excused if it was made one year or less before the effective filing date, and either (i) the disclosure was made by the inventor, a joint inventor, or someone who obtained the subject matter directly or indirectly from them, or (ii) the subject matter had already been publicly disclosed by the inventor before the intervening third-party disclosure. § 102(b)(2) provides a parallel exception for the § 102(a)(2) “secret prior art” category, including a common-ownership/joint-research-agreement exception relevant to university-industry collaborations. (MPEP § 2152 and MPEP § 2153 are USPTO’s authoritative walkthroughs of both exceptions.)
Two limits matter more in practice than the grace period’s existence:
- It is not a filing deadline extension, it is a narrow safe harbor. The grace period only protects against the inventor’s own disclosure (or a derivative one). If a third party independently discloses the same or obvious subject matter before the inventor’s own qualifying disclosure, that third-party disclosure is still prior art the grace period does not reach — waiting the full year on the assumption that “we’re protected” is a real risk, not a formality.
- On-sale activity counts even when it’s confidential. The Supreme Court held in Helsinn Healthcare S.A. v. Teva Pharmaceuticals USA, Inc., 586 U.S. 123 (2019) that a commercial sale to a third party who is contractually obligated to keep the invention confidential can still trigger the § 102(a) on-sale bar. For a TTO, this means a sponsored-research agreement, an option agreement, or an early licensing deal that includes a sale of the technology can start the one-year clock even though the deal itself is under NDA — “we kept it confidential” does not automatically avoid the bar.
Why disclosure timing is the real operational problem for a TTO
Nearly every public act a faculty inventor takes in the ordinary course of academic work can start the § 102(a)(1) clock: a conference presentation, a poster session, a thesis or dissertation deposited in a university repository, a journal article, a preprint, a grand-rounds talk, or a public demo at a sponsor site visit. None of these require the TTO’s knowledge or involvement to count as prior art against the TTO’s own eventual application. That asymmetry — the inventor doesn’t need permission to publish, but publishing can extinguish rights the TTO hasn’t yet perfected — is why invention-disclosure timing is treated as urgent internally, not administrative housekeeping.
The standard operational answer is: disclose to the TTO before the public disclosure, not after, so the office can decide whether to file a provisional application first. A provisional application is comparatively cheap, does not require formal claims, and establishes an effective filing date that then makes the researcher’s own subsequent publication fall inside, not outside, the AIA’s protected window. Filed after the public disclosure, the same provisional only starts the one-year US clock running from the disclosure date and does nothing for jurisdictions with no grace period at all.
How § 102 timing interacts with Bayh-Dole’s own deadlines
For inventions conceived or reduced to practice under federal funding, § 102’s novelty clock runs alongside — and is explicitly referenced by — the Bayh-Dole Act’s own disclosure and election-of-title deadlines under the standard patent rights clause at 37 CFR 401.14:
- The contractor (the university) must disclose each subject invention to the funding federal agency within two months after the inventor’s written disclosure to the institution’s patent-matters personnel.
- The university must then elect, in writing, whether to retain title within two years of that disclosure to the agency.
- Critically, 37 CFR 401.14(c)(2) lets the funding agency shorten the two-year election window whenever a public disclosure, sale, or other public availability has already started the § 102(b) one-year statutory bar running — the election deadline can be compressed to as little as 60 days before the end of that one-year period. In other words, an inventor’s own conference talk or paper doesn’t just risk the patent — under Bayh-Dole it can also collapse the university’s internal decision window from two years down to a matter of weeks.
This is the connective fact a generic patent-law explainer will not surface but a research-administration office needs: the § 102(b) public-disclosure clock and the Bayh-Dole election-of-title clock are not independent processes running on separate tracks — a disclosure event can shorten both simultaneously. See CASRAI’s federal grant closeout guide for how Bayh-Dole invention reporting fits into the broader closeout obligations under the same federal award.
A practical disclosure-to-filing timeline
| Event | Clock it starts or affects | Practical action |
|---|---|---|
| Inventor conceives/reduces invention to practice under a federal award | Starts the Bayh-Dole disclosure obligation | Inventor should disclose to the TTO’s patent-matters office promptly — the standard clause requires institutional disclosure to the funding agency within 2 months of the inventor’s internal disclosure |
| Any public disclosure (talk, poster, thesis, preprint, journal article, non-confidential sale, non-confidential public use) | Starts the § 102(a)(1)/(b)(1) one-year U.S. grace-period clock; simultaneously destroys absolute-novelty rights in most foreign jurisdictions immediately | File a provisional application before the disclosure wherever the timeline allows; if disclosure has already happened, file within the year and treat foreign filing as urgent |
| Confidential sale or license obligating the counterparty to secrecy | Can still start the § 102(a)(1) on-sale bar per Helsinn v. Teva | Do not assume an NDA removes the need to file before the transaction closes |
| University disclosure of the invention to the funding agency | Starts the Bayh-Dole 2-year election-of-title window (37 CFR 401.14(c)(2)) | Track election deadline in the TTO’s docketing system; confirm whether a prior public disclosure has shortened it to 60 days before the § 102(b) bar |
| U.S. provisional application filed | Establishes an effective filing date; starts the 12-month clock to file a non-provisional and/or a PCT application claiming priority | Calendar the 12-month PCT/foreign-filing deadline immediately on filing the provisional |
The international problem: absolute novelty has no grace period
The AIA’s one-year grace period is a U.S.-specific safe harbor. Most major foreign patent offices — including the European Patent Office and China’s CNIPA — apply an absolute novelty standard: any public disclosure of the invention anywhere in the world before the effective filing date destroys novelty for that jurisdiction, full stop, with only narrow exceptions (the EPO’s exception, for example, is limited to disclosures resulting from evident abuse or disclosure at certain officially recognized international exhibitions, and even that carries only a 6-month window). A disclosure that is entirely safe under § 102(b)(1) domestically can permanently foreclose patent protection in Europe, China, and most of the rest of the world the same day it happens. This is why TTOs treat “we’re still inside the U.S. one-year grace period” as reassurance about domestic rights only, never as a green light to delay filing if the institution has any interest in foreign patent protection.
Frequently asked questions
Does presenting research at a conference start the § 102 clock?
Yes. A conference talk, poster, or abstract that makes the invention “otherwise available to the public” is a disclosure under § 102(a)(1), regardless of whether a patent application has been filed. It starts the U.S. one-year grace-period clock and, in most cases, immediately eliminates novelty in absolute-novelty jurisdictions.
What’s the difference between § 102 (novelty) and § 103 (non-obviousness)?
§ 102 asks whether the exact claimed invention was already disclosed. § 103 asks a broader question: even if no single prior art reference discloses the whole invention, would the invention as a whole have been obvious to a person of ordinary skill in the field, in light of the prior art that does exist? An application can clear § 102 and still be rejected under § 103.
Does filing a provisional patent application “use up” the one-year grace period?
No — filing a provisional application is not a disclosure event under § 102(a)(1); it is a private filing with the USPTO, not a public disclosure, and it sets an effective filing date rather than starting the grace-period clock. Filing early is the standard way a TTO converts a looming public disclosure into a protected one.
Is a doctoral thesis deposited in a university repository a public disclosure?
It can be, once it is indexed and accessible to the public (not merely submitted internally for examination) — the specific triggering moment depends on institutional repository practices and has been litigated in specific fact patterns. TTOs generally treat the deposit/availability date, not the defense date, as the operative risk date and coordinate embargo timing with the graduate school where a patent filing is pending. See CASRAI’s dictionary entries on embargo and preprint for related publication-timing concepts.
Does Bayh-Dole require the university to publicly disclose federally funded inventions?
No — Bayh-Dole’s disclosure obligation runs from the university to the funding federal agency (via the iEdison system), not to the public. It is a reporting and title-election requirement, separate from, but timed relative to, any public disclosure the inventor makes that triggers § 102.
Related CASRAI resources
- Federal Grant Closeout: The Process and a Practical Checklist — Bayh-Dole invention reporting in the broader award-closeout context
- How to Choose a Preprint Server — preprints as a public-disclosure event research offices need to track
- Embargo — coordinating publication timing with patent filing
- Preprint — definition and disclosure implications
- Conflict of interest disclosure — related institutional disclosure obligations in research administration
This guide explains the general framework of 35 U.S.C. § 102 and related Bayh-Dole timing rules for research-administration audiences. It is not legal advice; invention-specific novelty, on-sale-bar, and foreign-filing determinations should be made with qualified patent counsel and the institution’s technology transfer office.
