Patent eligibility is broken.

The only semi-cogent arguments that I have ever heard in support of the status quo is that the U.S. Patent and Trademark Office issues too many broad, vague patents, and that 35 U.S.C. § 101 affords defendants a low-cost way of quickly invalidating bad-faith patent assertions on the pleadings.  The rest boil down to "I don't like patents."

While the former point has a grain of truth, in that the USPTO does grant patents that it should not from time to time, the law's current remedy based on Alice v. CLS Bank Int'l has the nuance of a high-energy shrapnel bomb.  Collateral damage has already impacted plenty of narrowly-scoped and useful inventions.  And the latter is equivalent to saying "I'm fine with bad laws being on the books as long as they help me."

To make an admittedly stark analogy, the Department of Justice has estimated that about 75% of violent crime in the U.S. is perpetrated by men.[1]  If a societal goal is to reduce violent crime, then incarcerating all men would cause the violent crime rate to drop dramatically and would eliminate the difficulty and cost of determining whether suspects were guilty or innocent through investigation, arrest, pleadings, trial, and so on.  Such a proposal is outrageous.  But not granting or invalidating certain types of patent matters on arbitrary grounds and with conclusory reasoning should generate at least a modicum of outrage.

In any event, both of the arguments above can be addressed in a more equitable and logical fashion with narrowly-tailored legislation.  Enter Senators Tillis and Coons.

This is the third time that one or both of the senators have proposed or introduced a bill to revise § 101.  The first two attempts garnered discussion but little tangible progress.  At this point, the patent community is feeling like Charlie Brown repeatedly getting the football yanked away.

Nonetheless, such proposals are worth of consideration, so let's take a look at the bipartisan Patent Eligibility Restoration Act of 2023.

First off, like previous proposals, the bill initially addresses definitions in § 100.  Notably:

(k) The term 'useful' means, with respect to an invention or discovery, that the invention or discovery has a specific and practical utility from the perspective of a person of ordinary skill in the art to which the invention or discovery pertains.

Cool.  A POSITA gets to decide what has "specific and practical utility".  Keeping that in mind, here is the meaty part (and I will focus on the impact to engineering and computing innovations, as we have already published an analysis from the life sciences perspective):

§ 101. Patent eligibility

(a) IN GENERAL.—Whoever invents or discovers any useful process, machine, manufacture, or composition of matter, or any useful improvement thereof, may obtain a patent therefor, subject only to the exclusions in subsection (b) and to the further conditions and requirements of this title.

(b) ELIGIBILITY EXCLUSIONS.—

(1) IN GENERAL.—Subject to paragraph (2), a person may not obtain a patent for any of the following, if claimed as such:

(A) A mathematical formula that is not part of a claimed invention in a category described in subsection (a).

(B)    (i) Subject to clause (ii), a process that is substantially economic, financial, business, social, cultural, or artistic, even though not less than 1 step in the process refers to a machine or manufacture.

(ii) The process described in clause (i) shall not be excluded from eligibility for a patent if the process cannot practically be performed without the use of a machine or manufacture.

(C) A process that—

(i) is a mental process performed solely in the human mind; or

(ii) occurs in nature wholly independent of, and prior to, any human activity.

(D) An unmodified human gene, as that gene exists in the human body.

(E) An unmodified natural material, as that material exists in nature.

(2) CONDITIONS.—For the purposes of subparagraphs (D) and (E) of paragraph (1), a human gene or natural material shall not be considered to be unmodified if the gene or material, as applicable, is—

(A) isolated, purified, enriched, or otherwise altered by human activity; or

(B) otherwise employed in a useful invention or discovery.

(c) ELIGIBILITY.—

(1) IN GENERAL.—In determining whether, under this section, a claimed invention is eligible for a patent, eligibility shall be determined—

(A) by considering the claimed invention as a whole and without discounting or disregarding any claim element; and

(B) without regard to—

(i) the manner in which the claimed invention was made;

(ii) whether a claim element is known, conventional, routine, or naturally occurring;

(iii) the state of the applicable art, as of the date on which the claimed invention is invented; or

(iv) any other consideration in section 102, 103, or 112.

(2) INFRINGEMENT ACTION.—

(A) IN GENERAL.—In an action brought for infringement under this title, the court, at any time, may determine whether an invention or discovery that is a subject of the action is eligible for a patent under this section, including on motion of a party when there are no genuine issues of material fact.

(B) LIMITED DISCOVERY.—With respect to a determination described in subparagraph (A), the court may consider limited discovery relevant only to the eligibility described in that subparagraph before ruling on a motion described in that subparagraph.

Now let's break this down into a few bite-sized chunks.

Part (b)(1)(A):  This provision seems to be stating that inclusion of or reference to mathematical formulas in claimed inventions does not impact eligibility, but a claim to a formula itself and in isolation would be ineligible.  Under that assumption, this language seems fine and would get us out of the current situation where the Supreme Court tells us in Alice that the claim of Diamond vs. Diehr is still eligible but gives us a test vague enough to invalidate that claim.

Part (b)(1)(B):  Here we have some problems.  Specifically, what is "a process that is substantially economic, financial, business, social, cultural, or artistic?"  How "substantially" does the process need to involve these excluded categories?  And does this provision only apply to process claims and not machine (e.g., device, system) or article of manufacture claims (e.g., computer readable media)?  As an example of the problems we can expect from use of vague terms like "artistic", what if my invention is an algorithm used by a digital camera to sharpen images?  The process might be technical but the outcome can be viewed as artistic.  Nonetheless, section (ii) seems to provide an out so long as the inventions can only be practically performed by computer.  This would positively impact patentees in the artificial intelligence and machine learning space where inventions often require training with large data sets.

Part (b)(1)(C):  This provision seems to shut the door on the mental process exception, one that is remarkably problematic in practice, not to mention completely illogical.  In short, some USPTO examiners and judges have taken the position that anything that could be done in theory by a human with the help of a computer or other tools, even if doing so would take millions of years, is an ineligible mental process.  If the legislation would make it such that a computer-implemented process is by definition not a mental process, then that is progress.

Part (c)(1):  This language is similar to that of Senator Tillis's proposed § 101 fix from last year, in that it is saying that "you need to consider the claim as a whole, and we really mean it".  Additionally, it reverses the ambiguity introduced when Alice and Mayo v. Prometheus misappropriated notions of §§ 102, 103, and 112 and shoved them into the § 101 inquiry.

Part (c)(2):  And here we have a bone for fans of the quick kill.  A court can, at any time and based on its own discretion, allow § 101 invalidity arguments with limited discovery.  Hopefully that is enough to keep some stakeholders happy.

So is this legislation perfect?  No.  There are still some ambiguities, mostly surrounding what we commonly refer to a business method inventions or hybrid business / technical inventions.  If litigated or left to USPTO interpretation, it is likely that we would end up right back where we are with the overly broad exception for "methods of organizing human activity" being used as a cudgel against claims with non-business merit.

But as they say, perfection is the enemy of good enough.  So it may be best to hold our noses and support this legislation, as it is does seem to be an incremental improvement over the current eligibility mess.

[1] https://bjs.ojp.gov/content/pub/pdf/cvus0702.pdf, table 38.