NEWS
Summary of Causam Enterprises, Inc. v. International Trade Commission (Fed. Cir. Oct. 15, 2025)
Wow, the Federal Circuit seriously narrowed the scope of some patent assignments.
The Federal Circuit held that a 2007 patent assignment covering “divisions, reissues, continuations, and extensions” did not include continuation-in-part (CIP) applications. The court based its decision on standard contract interpretation principles, emphasizing that it could not insert terms—like “continuations-in-part”—that the parties did not explicitly include.
The panel highlighted that “continuations” and “CIPs” have long been treated as distinct in patent law: continuations cannot add new matter, while CIPs can, and this difference has practical and legal consequences for patent ownership and recordation.
The opinion contrasts with University of New Mexico v. Knight (2003), where similar assignment language was held to include CIPs. The court distinguished Knight because that case involved additional agreements with broad language transferring ownership of “all related patents.”
While the panel’s analysis was clear, commentators noted it was somewhat formalistic, ignoring functional issues like potential “gamesmanship” — where a drafter might add nominal new material to refile as a CIP and thereby alter ownership.
Ultimately, the ownership dispute was moot because the related patent claim had already been found invalid for obviousness in a companion IPR proceeding.
Automated Telephone Line Open
We just opened up a new service: a telephone line: (949) 943-8311 , where you can call and get answers to basic Intellectual Property questions. We have text FAQs available on our website, and we have a text chatbot available too that uses a personalized Chat GPT to access information from the FAQs and Bob’s books: White Space Patenting, and Green Fields Patenting.
But if you want to ask the questions with voice rather than text them, this might be the option for you.
The new Voice line can be accessed at (949) 943-8311.
Discovery Gamesmanship Backfires: A Lesson in Playing Fast and Loose
Every so often, the Federal Circuit hands down a decision that’s less about patents and more about human nature. Magēmā (pronounced: Maggie-May) Technology v. Phillips 66 is one of those cases.
In the initial trial, Phillips convinced the magistrate during discovery that Magēmā didn’t need actual test data — formulas would be good enough. Then, once trial rolled around, Phillips told the jury the exact opposite: no test data, no infringement.
The district court let it slide, and the jury came back with a broad-brush “no infringement.” But on appeal, the Federal Circuit said, hold on a second. When a party spends the whole trial hammering on a theory it promised wasn’t necessary in discovery, you don’t get to call that “harmless error.” Judge Bumb, sitting by designation, had the perfect metaphor: this wasn’t pepper sprinkled here and there — the whole trial was marinated in it.
So, new trial. Phillips’ gamesmanship, instead of locking in a win, ended up boomeranging right back at them.
Magēmā also tried to raise a claim construction issue under O2 Micro, but that one fizzled. You can’t complain about claim scope being left to the jury if you never asked the court to construe the terms in the first place.
Because I hate litigation gamesmanship, I am very happy to see this outcome. Unfortunately it is allowed way too often. But at least this time, the party playing the games didn’t profit.
March-In Madness: Harvard, Bayh-Dole, and the Trump Administration
When politicians get involved in patent law, things usually get messy — and the latest spat between the Trump Administration and Harvard over the Bayh-Dole Act is no exception.
A Little Bayh-Dole Background
Back in the 1970s, the U.S. government had a problem: it was sitting on a mountain of patents — roughly 28,000 of them — generated from taxpayer-funded research. Fewer than 5% of those patents were ever licensed, let alone commercialized. Why? Because government ownership meant non-exclusive licenses, bureaucratic hurdles, and no incentive for anyone to put serious money into development.
Enter the Bayh-Dole Act of 1980. Congress flipped the script: universities, nonprofits, and small businesses could now keep the patents from federally funded research, provided they played by certain rules. It was a win-win: researchers got the chance to commercialize, investors got certainty, and the public (in theory) got products instead of dusty patent files.
But Congress also built in a big stick — the “march-in” rights. If an institution wasn’t doing its job, the government could “march in” and force licensing of the invention. The statute lists four triggers:
- No practical application.
- Health or safety needs not being met.
- Public use requirements not satisfied.
- Failure to substantially manufacture in the U.S.
Sounds dramatic, right? Except here’s the kicker: in over forty years, no administration has ever actually marched in. Plenty of petitions, lots of noise — zero follow-through.
Harvard in the Crosshairs
According to reports, the Trump Administration dusted off those march-in provisions and pointed them at Harvard. Commerce sent a letter accusing the university of failing to comply with Bayh-Dole. Harvard fired back, calling the move political payback for its defense of academic freedom.
Now, whether you buy Harvard’s explanation or the Administration’s, the bigger point is this: a march-in threat is almost always more about leverage than law. It’s the sword hanging on the wall. Agencies wave it around when they want to pressure a university or company to fall in line.
Here’s the practical takeaway for innovators: if you’re working with federal funding, Bayh-Dole compliance isn’t optional. You need to file invention disclosures on time, elect title properly, and make sure your commercialization plans check the right boxes. Screw up the paperwork, and you give an agency an opening — political or not — to come knocking.
And while Harvard can weather a march-in threat, a startup or small university spin-out might not. Investors don’t like uncertainty, and nothing says “uncertain” like the government hinting it could hand your patent to someone else.
The Trump–Harvard drama is a reminder that Bayh-Dole’s march-in rights are the patent world’s version of Bigfoot: everyone talks about them, nobody’s seen them in the wild, but the threat alone can be enough to change behavior.
Call of Duty Lawsuit Ends Without Attorneys’ Fees for Activision
The Fifth Circuit just weighed in on a copyright fight between retired wrestler Booker T. Huffman and video game giant Activision. Huffman claimed that Activision’s Call of Duty: Black Ops IV “Prophet” character infringed his G.I. Bro comic book poster. A jury sided with Activision, but the company wanted Huffman to foot the bill for its legal fees—arguing the case was frivolous.
The district court said no, and the Fifth Circuit backed that decision. Relying on the Supreme Court’s Fogerty factors, the court found Huffman’s claims weren’t frivolous or objectively unreasonable, noting the case touched on unsettled areas of copyright law. The panel stressed that fee awards aren’t automatic, and the trial judge’s detailed six-page analysis showed no abuse of discretion.
Not everyone agreed. In a sharp dissent, Judge Oldham called the suit “speculation piled on fantasy piled on a pipe dream” and said Activision should have been compensated for defending a baseless $32 million claim.
At the end of the day, Activision won the battle but had to eat its own legal fees—reminding us that “winning” in court doesn’t always mean walking away richer.
Secrets, Secrets Are No Longer Hard to Keep (in Court)
Previously, courts have required trade secret plaintiffs to identify their trade secrets with particularity before the defendants were required to respond to any discovery requests. In practice, no matter how a plaintiff would identify the trade secrets at issue, the defendant would refuse to respond to discovery requests on the grounds that the identification was insufficiently particularized. And the courts would refuse to intercede to state whether the identification was sufficient, or even what was needed to particularize the trade secrets. The net effect was that trade secret claims often went nowhere.
But this new case, Quintara Biosciences, Inc. v. Ruifeng Biztech, Inc., holds that trade secret claims under the Defend Trade Secrets Act (DTSA) do not require a plaintiff to identify its alleged trade secrets with “reasonable particularity” at the outset of litigation. Whether a plaintiff has sufficiently particularized a trade secret under the DTSA is usually a matter for summary judgment or trial.
This is different from the rules governing discovery in cases brought under California’s Uniform Trade Secrets Act (CUTSA)., which still requires a plaintiff to identify its alleged trade secrets with “reasonable particularity” at the outset of litigation.
You can read more about the case at this link:
Impact of Ninth Circuit’s Recent Decision in Federal Trade Secret Cases Under the DTSA – Calbar IP Section Newsstand – Powered by Lexology
AI’s secret ingredient: Text, Data, and Fair Use
Text and data mining (TDM) is basically grabbing lots of data from the internet. From that data, large language models and AI extract patterns and can generate text.
Accessing large amounts of text will, of course, very quickly start stepping on copyrighted toes. But there is a mechanism by which copyrighted materials don’t always need explicit permission for use: the Fair Use Doctrine.
In the US, many large tech companies rely on the fair use doctrine to access and justify training their AI systems.
Fair use allows the use of copyrighted materials in some instances. It’s never a blanket permission, it’s a case-by-case analysis based on four factors: 1) the purpose and character of the use, 2) the nature of the copyrighted work, 3) the amount and substantiality of the portion used, and 4) the effect on the market for the original work.
There are class action suits arguing that AI training is not fair use because it involves mass copying, but for the most part, courts have so far found the LLM’s training on copyrighted material ok, inasmuch as they are used for pattern recognition rather than reproducing the expressive content. But again, there are cases ongoing, so this may change, requiring some sort of larger scale licensing arrangements.