Patents

Pannu v. Iolab: The Three-Part Test for Who Counts as a Joint Inventor

The Federal Circuit set out the durable standard for joint inventorship and confirmed that misjoinder or nonjoinder is curable under Section 256 unless the patentee acted with deceptive intent.

September 8, 2025
Close-up of an intraocular lens implant held with forceps
An improved intraocular lens prompted the test courts still use to decide who is a joint inventor. Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

Pannu v. Iolab Corp., 155 F.3d 1344, Nos. 97-1466, 97-1501 (Fed. Cir. Aug. 6, 1998), supplies the analytical spine of modern joint-inventorship law. While Ethicon showed what happens when a co-inventor is left off a patent, Pannu answered the prior question: what does it actually take to be a joint inventor in the first place? The court articulated a three-factor test that has been quoted in countless opinions since — and recently invoked by the USPTO in its guidance on AI-assisted inventions. Pannu also clarified the relationship between inventorship error and validity: a patent with the wrong inventors is not automatically void, because 35 U.S.C. § 256 allows correction unless the error was the product of deceptive intent.

At a glance

  • Case: Pannu v. Iolab Corp., 155 F.3d 1344 (Fed. Cir. 1998)
  • Docket: Nos. 97-1466, 97-1501; decided August 6, 1998
  • Court: U.S. Court of Appeals for the Federal Circuit
  • Panel: Chief Judge Mayer, Judge Lourie, Judge Rader
  • Patent: U.S. Reissue Patent No. 32,525 — an improved intraocular lens designed to reduce tissue snagging during cataract implantation
  • Parties: Dr. Jaswant S. Pannu (named inventor/plaintiff); Iolab Corp. (accused infringer)
  • Holdings: (1) the three-part test for joint inventorship; (2) misjoinder/nonjoinder is not an automatic invalidity ground because § 256 permits correction absent deceptive intent
  • Status: Vacated and remanded in relevant part for the district court to apply the correct framework

The dispute over a co-inventor

Dr. Jaswant Pannu held Reissue Patent No. 32,525 covering an improved intraocular lens — an artificial lens implanted after cataract surgery — engineered so that its supporting elements would not snag delicate eye tissue during insertion. Pannu sued Iolab for infringement. At trial, evidence emerged that another person, William Link, had contributed to the invention, yet Link was not named on the patent. Iolab argued that the omission rendered the patent invalid for improper inventorship.

The district court’s handling of that argument is what brought the case up on appeal. The court suggested that even if Link were a co-inventor, the patent could be invalidated only on a showing that Pannu had acted in bad faith in failing to name him — and, finding no such evidence, it treated the inventorship as essentially correctable and let a jury verdict stand. The Federal Circuit found the analysis muddled and used the appeal to lay down the proper two-step framework: first decide whether the omitted person is actually a joint inventor, and only then ask whether the resulting error is correctable.

The three-part joint-inventorship test

The court’s most enduring contribution is its statement of what makes someone a joint inventor. Conception is the touchstone, but not every contribution to a project rises to the level of co-invention. To qualify, a putative inventor must:

  1. Contribute in some significant manner to the conception or reduction to practice of the invention.
  2. Make a contribution to the claimed invention that is not insignificant in quality, when that contribution is measured against the dimension of the full invention.
  3. Do more than merely explain to the real inventors well-known concepts and/or the current state of the art.

The test deliberately filters out two categories of participants who often feel they deserve credit but legally do not: those who supply only ordinary skill or routine assistance in reducing an idea to practice, and those who merely convey the existing state of the art or well-understood principles. A contribution must be both qualitatively significant relative to the whole invention and genuinely inventive rather than instructional. The framework’s flexibility — “measured against the dimension of the full invention” — is precisely what has made it adaptable, and it is why the USPTO turned to the Pannu factors when assessing whether a natural person made a significant enough contribution to an AI-assisted invention to be a proper inventor.

Section 256 and the deceptive-intent gate

The second holding addresses what follows once a court finds the named inventors wrong. Improper inventorship — naming too many inventors (misjoinder) or too few (nonjoinder) — is a condition of patentability under 35 U.S.C. § 102(f) (pre-AIA). But the Federal Circuit emphasized that such an error does not automatically doom the patent, because Congress provided a cure. Section 256 allows a court to order correction of inventorship when, “through error,” a person is named who should not be, or an inventor is omitted who should be named — provided the error arose without deceptive intention.

The court therefore prescribed a sequence. First, the factfinder determines whether the alleged omitted (or wrongly named) person is in fact a joint inventor under the three-part test. If so, the court must then consider whether § 256 can save the patent: it can, unless the party seeking to invalidate proves by clear and convincing evidence that the inventorship error resulted from deceptive intent. Good faith is presumed; the patentee does not bear the initial burden of disproving deception. Only an uncorrectable error — one infected by deceptive intent — invalidates the patent. Because the district court had collapsed these inquiries, the Federal Circuit vacated and remanded for proper application of the framework.

Open questions

Pannu gives a test but not a formula, and its qualitative language leaves recurring uncertainty. How “significant” a contribution must be, and how courts should weigh it “against the dimension of the full invention,” are intensely fact-bound questions that continue to divide litigants — particularly for collaborators who contribute to some claims but not others, or whose input straddles the line between inventive insight and skilled execution. The deceptive-intent standard has its own afterlife: the AIA’s amendments reshaped the statutory landscape around inventorship and removed § 102(f), and although § 256 correction survives, the precise role of “deceptive intent” and the interaction with inequitable-conduct doctrine after Therasense remain contested. Finally, the application of the Pannu factors to machine-assisted invention — where the “contributor” is a tool rather than a person — is only beginning to be worked out.

Implications

  • Use the three factors as a checklist. Before filing, evaluate each contributor against significance of contribution, quality relative to the whole invention, and whether they went beyond explaining known art.
  • Distinguish inventors from helpers. Supplying ordinary skill, routine reduction to practice, or a recital of the state of the art does not make someone a joint inventor.
  • Inventorship errors are usually fixable. Misjoinder and nonjoinder do not automatically invalidate a patent; § 256 permits correction absent deceptive intent, with good faith presumed.
  • Challengers must prove deception to invalidate on inventorship. An accused infringer attacking a patent for wrong inventorship must show, by clear and convincing evidence, that the error stemmed from deceptive intent.
  • Document contributions contemporaneously. Inventor declarations, lab records, and contribution logs make both the Pannu analysis and any later § 256 correction far more defensible.

Frequently asked questions

What is the Pannu test? It is the Federal Circuit’s three-part standard for joint inventorship: a co-inventor must (1) contribute significantly to conception or reduction to practice, (2) make a contribution not insignificant in quality measured against the full invention, and (3) do more than explain well-known concepts or the state of the art.

Does naming the wrong inventors automatically invalidate a patent? No. Under Pannu, improper inventorship is correctable under 35 U.S.C. § 256 unless the challenger proves the error resulted from deceptive intent. Good faith is presumed, so an innocent mistake can be fixed rather than fatal.

Why does Pannu matter for AI-assisted inventions? The USPTO has invoked the Pannu factors to assess whether a natural person contributed enough to an AI-assisted invention to be a proper named inventor — a person must make a significant contribution under these factors, not merely oversee or prompt an AI system.

Authorities and sources