Remarks by Director Squires at the 2025 AIPLA Annual Meeting

Remarks as written

John A. Squires, Under Secretary of Commerce for Intellectual Property and Director of the USPTO

American Intellectual Property Law Association Annual Meeting

October 31, 2025

Thank you, Vince, for that kind and generous introduction.

It is an honor to be here with you-a community of counselors, innovators, advocates, and entrepreneurs-to share not only where the United States Patent and Trademark Office now stands today, but also perhaps more importantly, where we are going-together.

And before I begin, let me thank you again for recognizing our USPTO superheroes. Because that's what they are-they certainly don't need a Halloween costume to prove it. They prove it every day to make America's Innovation Agency the greatest in the world, because they are the ones who make everything in American intellectual property possible. Our examiners, our supervisors, our attorneys, and all of those across the enterprise who support them are the best in the world at what they do. Bar none. They are immensely deserving of our gratitude, and I'm so pleased that you honored them so well today.

AIPLA has supported this Office for well over a century. As we build a new chapter of cooperation, clarity, and confidence, I look forward to working with you closely in the months and years ahead. It is truly an honor to be here.

Today, I am wrapping up week five as Under Secretary of Commerce for Intellectual Property and the 60th Director of the United States Patent and Trademark Office. And as 60th Director, I have to tell you, I did receive a very warm welcome from the Government in my first week-it shut down. The good news? As you may know, we are a user-fee funded agency, and we are "open for business" as I've said time and again, operating on reserves-enough to power us comfortably through the end of the year. And then some, if needed.

The USPTO is the running engine of American ingenuity-and a shutdown does not mean we are stuck in idle. No way. We are working in overdrive as you've no doubt seen from our relentless drumbeat of new announcements. American innovation does not pause. Neither do we. With our reserves just over $1 billion total, we even offered to help pay America's heroes, our troops. But for reasons that I'll never understand, having spent my career in the private sector, I was told we can't do that. But I ask again, every single week. We will continue to strive to be a stabilizing force in uncertain times. And a force in overdrive.

In just five weeks, we've:

• Begun a new era of eligibility with the signing the first two patents of my tenure, a crypto/DL business method and a medical diagnostic.
• Closed an unprofitable regional Outreach Office.
• Returned the agency to its core mission by executing a long planned for reduction in force.
• Launched ASAP!-our AI-assisted Automated Search Pilot Program.
• Issued half a dozen Director Review decisions, addressing issues including unfocused petitions and inconsistent expert testimony.
• Presided over 50 institution decisions.
• Issued the second-ever ARP decision in Desjardins, supporting the eligibility of AI-innovations.
• Created a Director/Employee feedback mechanism.
• Delivered a comprehensive written Statement for the Record on Patent Eligibility to the U.S. Senate.
• Hosted the 2025 Collegiate Inventors Competition.
• Held dozens of stakeholder meetings, of which AIPLA was among the very first.
• Issued a Notice of Proposed Rulemaking on "one and done" validity challenges.
• Restored the original intent of the AIA with Director Institution of AIA Proceedings.
• Hosted the 43rd Annual Meeting of the Trilateral Offices, with the JPO, EPO, and WIPO as an observer.
• Announced the Trilateral AI Working Group.
• Announced Streamlined Claim Set Pilot Program.
• Hosted the 11th Annual ID5 meeting, with Korea, China, Japan, the EPO, and WIPO as an observer.
• Hosted the 2025 TM5 meeting, also with Korea, China, Japan, the EPO, and WIPO as an observer.
• Launched our new community engagement model and solicited comments for locations formerly served by the Rocky Mountain Regional Office.

And those are just the highlights against the backdrop of our day-to-day responsibilities. That pace alone is remarkable. But I will take a moment to tell you just how remarkable by taking you back to week minus-forty. On January 20th, we inherited an unexamined patent application backlog that was an absolute dumpster fire. "Unacceptable," as Secretary Howard Lutnick rightly termed it.

• 837,928 unexamined applications-an all-time, sad record.
• The backlog jumped from 576,103 in 2020 to 837,928 in January 2025-an increase of 261,825 unexamined applications. I give you the exact numbers because every application counts, every application matters. What if it was your application?
• Yet annual filings fell by more than 8,000 over the same period.

A total betrayal of American inventors who deserve better. And a betrayal of our Examiners who are always there to meet the moment. But, from the second she was on the job, Deputy Director Coke Stewart would have none of it. And we went back to our core mission then and there. Examiners were brought back to actual examination. We reined in travel, returning more than 1,200 workdays per month. We reevaluated all of our contracts. These actions resulted in $315 million in savings, a full one-third of the reserves we're now using to stay open and almost a twice what I need to keep the Trademark Business Unit open into the new year. But that is then. This is now.

Now, we are in overdrive. We are replacing that wasted time, money, and opportunity with:

• 1,100 new examiner hires under way.
• New incentives and bonuses.
• And new IT tools-including AI-enhanced search-to help our examiners find the needle hidden in the global prior-art haystack. And apply it. Up front. So, our patents can be "born strong."
• And just this week, we launched the Streamlined Claim Set Pilot Program to cut pendency even further. This is another tool in our arsenal to speed things up.
• Two weeks ago, we launched our AI-assisted Automated Search Pilot, called ASAP!, so you get the prior art right now when it does the most good. AI helped Trademarks profoundly. The Trademarks Register Protection Office removed around 61,00 invalid application and registrations clogging the pipes for good faith applications.
• But most notably, as of last month, our amazing Patents team has beaten the backlog down big-time-a reduction of nearly 50,000 backlogged applications in just over 6 months-with rising quality.
• And our trademarks team reports their lowest pendency ever-again with rising quality.

By no means am I declaring victory. But I am declaring we are not going back-ever. Instead, we are doubling down-and with AI, we're going to re-invest in our examiners-who as you heard me then and believe me now, because you've seen their breaking-the-back-of-the-backlog heroics-are the best in the world at what they do.

So, turning to present day and our bright future-I'm delighted to steward the steamship U.S.S. USPTO full speed ahead. And what's in our engine room? AI.

AI is the Roy Kent of modern technology-it's here, it's there, it's everywhere.

In all seriousness, there is but one, singular, undeniable, irresistible force that unites us as innovators and the Office: AI. I would be unforgivably remiss in my first Director's address if I didn't devote these remarks to the most transcendent and transformative technology of our time, perhaps of any time. In no way, shape, or form can we have a discussion about the future of intellectual property without AI. In no way, shape, or form can America have any credibility-domestically or on the world stage-if we don't meet the moment. These are the very types of moments the Founders foresaw in enshrining exclusive rights in our Constitution. As such, let me begin with three main areas where we must address as to AI, or it will steamroll us as it hitchhikes through the universe.

• The first is with respect to our operations and the charge given to our agency.
• The second, AI as a technological revolution.
• And third, technological revolutions and what they've taught us about eligibility.

AI and our operations

So, AI as to our Office and operations. I will give you a line of sight about our leadership and the way we look at patents. Between our Secretary-a prolific inventor as you know with over 400 patents, our Deputy Secretary, and yours truly, we have collectively almost a century in Wall Street, start-up, and technology experience. So, there's a throughline to our work, and it is that patents are soft dollar assets. Soft power. The linkage between rights and real economy. And with a big economic stick. This is why you've heard me term the USPTO as the Department of Commerce's Central Bank of Innovation. Every piece of intellectual property we put into circulation is a potential new job, a new business, a competitive advantage, an investible asset.

Furthering this notion, at that edge of innovation, then is the question of what to make of these new, useful, and non-obvious ideas? I invoked it at my confirmation hearing, the adage, "every patent begins its life as a trade secret." Let me refine that ever so slightly. Every patent begins its life as a trade. Disclosure for exclusivity. That's the constitutional bargain of course. That means this Office and applicants are potential counterparties to a trade, not adversaries in a contested litigation. That also means our world-class examiners are in fact chief negotiators on behalf of the United States of America-finding the line where prior art leaves off and inventive contribution begins.

If price discovery is the difference between bid and ask, on the market floor of the Central Bank of Innovation, this is claims discovery. The knife's edge of disclosure for exclusivity. Where the trade is made. And the best trades, the deals to be made, occur when both sides have reliable information. The better the disclosure, the better the trade.

And now, in our operations, in our first pilot, that is where AI comes in. We unveiled last month-and launched it last week our ASAP search pilot. ASAP! gets you the prior art right now-before the first Office Action-so applicant and examiner are literally on the same page in their trade decision. We also just announced streamlined claim sets, which you can think about as structured trades. That simple shift-transparency at the start-improves quality across the board.

With AI, there's no excuse for missing prior art. Machines can surface it faster, broader, and smarter than ever before. That makes the trade cleaner, the issuance stronger, and the asset more investible. And we are open for business.

AI as technological revolution

Turning next to AI as technological revolution. There was a similar technological revolution that saved New York City in the early 1900s. The biggest problem New York City faced in the late 1800s was-horse manure. Experts predicted that within decades, the city would drown in it. Then, the horseless carriage was invented. Ten years later, the problem vanished.

That's bypass: when technology just blows through a problem, delivers us to a new era, and everything gets re-engineered around it. So, what does this re-engineering look like? It looks like the dot-com boom 25+ years ago. It looks like the world that Satoshi Nakamoto foresaw with distributed ledger 15+ years ago. And it looks a heck of a lot like the AI revolution we're witnessing now. But for inventors and brand owners-intellectual property, writ large-what does this look like in the real world?

Let's take distributed ledger technology-DL-for short. And let's examine, in the crucible of innovation, what is DL technology good at? Trust. Trust enabled by a structured data schema-the ledger. And what's it not so good at? Speed.

Let's turn to AI. What's it good at? Speed. Speed algorithmically to generate predictive outcomes with respect to massive amounts of unstructured data. LLMs. What's it not so good at? Trust.

So, the inventive impetus-the omnibus and pervasive re-engineering we are seeing-is based, very generally, but profoundly on the presence or absence of trust. I saw this firsthand because that's how I obtained Goldman's first blockchain patent- SETLCoin. SETLCoin allows a securities trade to be settled and cleared instantly when the bid is hit. There's no need for a central party to provide trust-the ledger does it. So, no need for DTCC. Note that DL's been around for 15+ years-but the re-engineering is ongoing. With upwards of 10,000 issued crypto patents and 30,000 applications at last count to prove it.

What about AI? The re-engineering centers around affirmatively providing trust, or structure, or some component of interaction. DL because of trust cuts out central counterparties, AI because it's lacking in trust, yearns for it. So, when I say AI is everywhere- and you can blame Roy Kent or not-I don't just mean the headlines or the chatbots. I mean it's reshaping the architecture beneath: how data is processed, what improvements are made, how rights are traded, how data and arrays are structured and accessed. That's not hype-that's infrastructure. And that's invention. That's what we do. And we are open for business.

When you say infrastructure and re-engineering-like the internet 25-30 years ago, like DL 15-20 years ago, like AI here and now-the conversation turns to patents. I see eligibility. You could say I'm like the boy in the movie the Sixth Sense who sees dead people; I see eligibility. And on Halloween, I'm here to tell you and show you-I'm not seeing ghosts. Why?

Patent eligibility

Because we have an existing tried, true, and trusted schema to discern patent eligibility that's been right here all along. Hiding in plain sight. Under 35 U.S.C. § 101, Congress gave us broad terms we well understand: process, machine, manufacture, composition of matter. Congress chose its words carefully because it knew we couldn't predict the revolutions to come. We just never know when horseless carriages will make a societal woe the punchline of the future.

Indeed, courts have rightly carved out narrow exceptions-laws of nature, physical phenomena, abstract ideas-but that's all they are: narrow. We have the divining rod to sift between eligibility and patentability-and we've had it all along. There are actually three. And if we allow these three pillars to do the work on eligibility, then as Congress designed, sections 102, 103, and 112 can do the work on patentability.

What are the 3 pillars of eligibility? In combination, they are:

First, 35 U.S.C. § 100(b), which provides a definition of "process" encompassing specifically new uses of known technologies-and it has done so since the Patent Act of 1793. Think about that.

Second, the Enfish decision, which holds as patent eligible improvements to computer data structures. DL, if nothing else, is the ultimate data structure-and a technology that's been around for more than 15 years. And re-engineering is therefore new uses under 100(b). And AI, claimed and supported correctly, is merely the other side of the Enfish coin.

In Ex parte Desjardins, the claim related to improvements to machine learning models to "effectively learn new tasks in succession whilst protecting knowledge about previous tasks." But this had been abstracted out of existence sua sponte by the Desjardins Board-trying to do the right thing, mind you. And I am by no means picking on them whatsoever. The guidance simply hasn't been there. But believe me it will be shortly. Eligibility simply can't be relegated to the eyes of the beholder. And it will not be on my watch.

If there is any one thing I am going accomplish in my tenure, it's this-making sure the door to the patent office is wide open to transformative technologies. Bottom line, for eligibility-you have to know what to be looking for. So, what is it after all this that we are looking for??

That's the third pillar: something more. The something more that Alice and Mayo tell us to look for. The artifacts of re-engineering that revolutionary technologies provide. The improvements, the application of technology. When the system changes the architecture itself-how information flows, not just what it does. That's why AI and distributed-ledger innovations aren't fringe. They are the center of what eligibility was written for. Eligibility follows architecture. When technology rewires the foundation, when everything around it is re-engineered, that's when invention begins.

But we don't have to look ahead and be confused. We can look back and be confident. Where? Look no further than Samuel Morse's Claim 5-the dot and dash system he devised to transmit the alphabet electromagnetically over a wire. Whereas Claim 8 recited simply electromagnetism. Unpatentable as a law of nature-lacking something more. Lacking practical application. Lacking eligibility.

When technology surprises us, as it does every so often, we are not unguided and alone on technology's frontiers. We need only to look back at one of the most transformative, famous, and lifechanging technologies the world has ever seen. Samuel Morse's telegraph and the way he devised-not the patent eligible way he devised it-to transmit data and information over wires. And 171 years later, Alice and Mayo cannot be reasonably read as overruling it.

The eligibility of Claim 5 against the backdrop of Claim 8 is still good law. So, something more? I'll see you and raise you something more-how about something Morse. Now you know exactly why I made a point of signing the first two patents I did.

• U.S. Patent No. 12,419,201 for Materials and Methods for Performing Histochemical Assays For Human Pro-Epiregulin and Amiphiregulin; and
• U.S. Patent No. 12,419,202 for Systems and Methods For Generating An Architecture For Production of Goods and Services.

They conferred not only something more to be eligible, but also were tested by the fires of sections 102, 103 and 112-and passed. And now you see the eligibility I saw in Desjardins. Not a dead man walking.

So, when it comes to the AI revolution, I give you the trilogy of section 100(b), Enfish, and something more, something Morse. And with that, I leave you with this.

As to the technological revolution de jour, today it's AI. Tomorrow, who knows what form this amazing shape shifter we call innovation will take.

The doors to the USPTO-let me restate that-the doors to America's Innovation Agency are wide open. Even when no one else's are. We are open for business. Please come.

Thank you.