USPTO Director Reclaims Authority to Institute IPRs

The new Under Secretary of Commerce for Intellectual Property and Director of the United States Patent and Trademark Office, John Squires, has reclaimed authority to institute IPRs. Here is his letter regarding the subject:

An Open Letter From America’s Innovation Agency

Bringing the USPTO Back to the Future: Return of Institution Authority under 35 U.S.C. §§ 314 and 324 to the Director

Dear Colleagues, Inventors, and Americans,

Under the America Invents Act (AIA), Congress entrusted the United States Patent and Trademark Office with several mandates to ensure the timely and fair adjudication of patent validity challenges through post-grant review (PGR) or inter partes review (IPR) mechanisms and priority contests via derivation proceedings. As to IPRs specifically, under 35 U.S.C. § 314(a), Congress made plain that: The Director may not authorize an inter partes review to be instituted unless the Director determines that the information presented in the petition … shows that there is a reasonable likelihood that the petitioner would prevail with respect to at least 1 of the claims challenged in the petition. This statutory language expressly vests the authority to institute IPRs and PGRs in the USPTO Director. While 35 U.S.C. § 3(b)(3)(B) permits delegation of that authority, such delegation is non-exclusive. Statutorily, the Director retains full and concurrent authority over whether an IPR or PGR shall proceed. Since the AIA’s enactment, initial operational choices led to the delegation of institution decisions to the Patent Trial and Appeal Board, where panels then adjudicated the merits once instituted. Although this delegation was initially practical, experience has raised structural, perceptual, and procedural concerns inconsistent with the AIA’s design, clear language, and intent affecting, among other things, the public’s rightful expectation of impartiality. Given the statutory charge, my aim as Director is to address these concerns. Under oath in my confirmation hearing before the Senate Judiciary Committee and thereafter in my submitted Questions for the Record responses, I expressed discomfort that data seemed to be “skewed” in favor of certain provisions (namely IPRs over PGRs and a very high invalidation rate). To me, this raised questions about both the administration of IPR proceedings and their institution in particular. I vowed to administer the AIA as the statute provides and as Congress intended.

Today, in keeping with my vow and having now taken the Oath of Office as USPTO Director, I have ordered changes pursuant to my memo to the Board (attached). Below, I describe the reasons for my action today. Over the past several years, the delegated-institution model has given rise to the following difficulties: 1. Perception of Self-Incentivization – While the Board has done an admirable job, performance metrics and workload structures have created the appearance that institution decisions affect docket size, credit, and resource allocation—inviting concern that the Board may be “filling its own docket.” – This may be unfounded, but nevertheless such a perception undermines public confidence in the integrity of our Office’s adjudicatory functions with respect to IPRs. 2. Bifurcated Procedures for Discretionary Considerations – The evolution of the bifurcated processes, which were smart and necessary, was never intended to be permanent. Under those processes, a preliminary review precedes Board referral. However, this appears to have inadvertently produced extraordinarily high institution rates (at one point exceeding 95 percent) for referred cases. 3. Statutory Adherence and Administrative Clarity – Congress expressly charges the Director—not the Board as delegees —to make institution determinations. Returning this function to the Director re aligns our Office’s procedures with the clear language and intent of the statute and returns accountability for such decisions to the Director just as the framework of the AIA provides. In sum, reclaiming the Director’s statutory role is intended to: • Eliminate the appearance of self-interest by separating the power to institute from the body that conducts the trial; • Remove a perceived referral-signal bias by centralizing the decision point; • Enhance transparency and public trust through a single line of authority; and • Re-align the duties and responsibilities of the Director, as a Presidentially appointed and Senate-confirmed officer, to be accountable for this threshold determination and properly effectuate the clear language of the AIA and thus Congress’s intent.

This action aligns the USPTO’s administration of IPRs with both the letter and the spirit of 35 U.S.C. § 314 and strengthens the integrity of the Office’s adjudicatory processes. In closing, the mission of America’s Innovation Agency is to lead the world in intellectual property protection. We can do so and serve the public interest only by maintaining a patent system that is fair, predictable, and respected. Returning institution authority to the Director bolsters our mission because it restores the statutory framework mandated by Congress in the America Invents Act.

UK intervention on SEPs including rate setting swims against the tide

The UK Intellectual Property Office’s 2025 consultation on standard essential patents proposes measures to improve licensing transparency and efficiency. These include searchable SEP databases, essentiality checking services, and mechanisms for aggregate rate setting to facilitate top-down approach FRAND licensing rate apportionments. While these initiatives aim to support UK innovation – particularly for SMEs – they risk undermining a licensing system that has successfully evolved through decades of commercial practice and judicial developments.

The UK’s Prime Minister and Chancellor have recently pledged to promote economic growth by slashing red tape and taking out regulators. The IPO’s proposals fly in the face of that.

In my response submission to the IPO’s consultation, I focus on interventions that could do more harm than good: essentiality checking and essentiality rate estimating, aggregate royalty setting and top-down rate apportionment. My concerns are similar to those previously raised in response to the EU SEP consultation in 2023 and 2024.

Checking essentiality – along with infringement and validity – are important and are economically achieved on handfuls of patents to reliably establish that licensing is required. Comparable licences are then the generally preferred method of determining FRAND rates. Where these do not yet exist or are unavailable, parties are best placed to determine rates through discussion and negotiation.

Estimating essentiality rates of entire patent portfolios and for all patents reading on a standard is a far more demanding and costly endeavour, even when only random samples of patents are checked. Results are inaccurate and unreliable.

Setting aggregate royalties and then apportioning them based on counts of declared essential or checked essential patents is also very problematic.

My consultation submission can be downloaded here.

U.S. NIST Report on DeepSeek

The U.S. National Institute of Standards and Technology has recently released a 69 page report concerning DeepSeek and the risks associated with utilizing it.  The Press Release states:

WASHINGTON — The Center for AI Standards and Innovation (CAISI) at the Department of Commerce’s National Institute of Standards and Technology (NIST) evaluated AI models from the People’s Republic of China (PRC) developer DeepSeek and found they lag behind U.S. models in performance, cost, security and adoption.

“Thanks to President Trump’s AI Action Plan, the Department of Commerce and NIST’s Center for AI Standards and Innovation have released a groundbreaking evaluation of American vs. adversary AI,” said Secretary of Commerce Howard Lutnick. “The report is clear that American AI dominates, with DeepSeek trailing far behind. This weakness isn’t just technical. It shows why relying on foreign AI is dangerous and shortsighted. By setting the standards, driving innovation, and keeping America secure, the Department of Commerce will ensure continued U.S. leadership in AI.”

The CAISI evaluation also notes that the DeepSeek models’ shortcomings related to security and censorship of model responses may pose a risk to application developers, consumers and U.S. national security. Despite these risks, DeepSeek is a leading developer and has contributed to a rapid increase in the global use of models from the PRC.

CAISI’s experts evaluated three DeepSeek models (R1, R1-0528 and V3.1) and four U.S. models (OpenAI’s GPT-5, GPT-5-mini and gpt-oss and Anthropic’s Opus 4) across 19 benchmarks spanning a range of domains. These evaluations include state-of-the-art public benchmarks as well as private benchmarks built by CAISI in partnership with academic institutions and other federal agencies.

The evaluation from CAISI responds to President Donald Trump’s America’s AI Action Plan, which directs CAISI to conduct research and publish evaluations of frontier models from the PRC. CAISI is also tasked with assessing: the capabilities of U.S. and adversary AI systems; the adoption of foreign AI systems; the state of international AI competition; and potential security vulnerabilities and malign foreign influence arising from the use of adversaries’ AI systems.

CAISI serves as industry’s primary point of contact within the U.S. government to facilitate testing, collaborative research, and best practice development related to commercial AI systems, and is a key element in NIST’s efforts to secure and advance American leadership in AI.

Key Findings

DeepSeek performance lags behind the best U.S. reference models.
The best U.S. model outperforms the best DeepSeek model (DeepSeek V3.1) across almost every benchmark. The gap is largest for software engineering and cyber tasks, where the best U.S. model evaluated solves over 20% more tasks than the best DeepSeek model.

DeepSeek models cost more to use than comparable U.S. models.
One U.S. reference model costs 35% less on average than the best DeepSeek model to perform at a similar level across all 13 performance benchmarks tested.

DeepSeek models are far more susceptible to agent hijacking attacks than frontier U.S. models.
Agents based on DeepSeek’s most secure model (R1-0528) were, on average, 12 times more likely than evaluated U.S. frontier models to follow malicious instructions designed to derail them from user tasks. Hijacked agents sent phishing emails, downloaded and ran malware, and exfiltrated user login credentials, all in a simulated environment.

DeepSeek models are far more susceptible to jailbreaking attacks than U.S. models.
DeepSeek’s most secure model (R1-0528) responded to 94% of overtly malicious requests when a common jailbreaking technique was used, compared with 8% of requests for U.S. reference models.

DeepSeek models advance Chinese Communist Party (CCP) narratives.
DeepSeek models echoed four times as many inaccurate and misleading CCP narratives as U.S. reference models did.

Adoption of PRC models has greatly increased since DeepSeek R1 was released.
The release of DeepSeek R1 has driven adoption of PRC models across the AI ecosystem. Downloads of DeepSeek models on model-sharing platforms have increased nearly 1,000% since January 2025.

A Good Idea to Cut Public Funding for R&D? Doesn't Look Like It.

David Rotman has authored an excellent article titled, “How to Measure the Returns on R&D Spending: Forget the Glorious Successes of Past Breakthroughs—the Real Justification for Research Investment is What We Get for Our Money. Here’s What Economists Say” in the MIT Technology Review.  The article does a nice job explaining recent research concerning public funding of R&D and the return on investment to the public.  The article notes that many questions are not answered.  The article is available, here. 

Medicaid to Receive Most Favored Nation Pricing on Pfizer Drugs

The White House has announced that President Trump has made a deal with Pfizer for state Medicaid to receive most favored nation pricing on Pfizer pharmaceuticals.  This is welcome news; however, I do wonder if the rest of the world’s pharmaceutical prices will rise.  The White House has released a fact sheet that states:

ADVANCING MOST-FAVORED-NATION PRICING: Today, President Donald J. Trump announced the first agreement with a major pharmaceutical company, Pfizer, to bring American drug prices in line with the lowest paid by other developed nations (known as the most-favored-nation, or MFN, price).

• The agreement will provide every State Medicaid program in the country access to MFN drug prices on Pfizer products, resulting in many millions of dollars in savings and continuing President Trump’s historic efforts to strengthen the program for the most vulnerable.
• The agreement ensures foreign nations can no longer use price controls to freeride on American innovation by guaranteeing MFN prices on all new innovative medicines Pfizer brings to market. 
• The agreement requires Pfizer to repatriate increased foreign revenue on existing products that Pfizer realizes as a result of the President’s strong America First U.S. trade policies for the benefit of American patients. 
• The agreement requires Pfizer to offer medicines at a deep discount off the list price when selling directly to American patients.

 
DELIVERING REDUCED COSTS: Today’s actions will result in tangible cost savings to American patients and the healthcare system as a whole. Taken together, more than 100 million patients are impacted by the diseases Pfizer’s medicines treat, and many of those will benefit from the President’s successful negotiation of lower prices for Americans. Examples include:

• Eucrisa, a topical ointment for atopic dermatitis, will be made available at an 80% discount to patients purchasing directly.
• Xeljanz, a widely used oral medication for rheumatoid arthritis, psoriatic arthritis, and ulcerative colitis, will be available at a 40% discount to patients purchasing directly.
•  Zavzpret, a commonly utilized treatment for migraines, will be sold directly to patients at a 50% discount.

 
ENDING GLOBAL FREELOADING ON AMERICAN PHARMACEUTICAL INNOVATION:  President Trump is taking decisive action to rebalance a system that allows pharmaceutical manufacturers to offer low prices to other wealthy nations while charging Americans significantly higher prices.  

• According to recent data, the prices Americans pay for brand-name drugs are more than three times the price other Organization for Economic Cooperation and Development nations pay, even after accounting for discounts manufacturers provide in the U.S. 
• The United States has less than five percent of the world’s population, yet roughly 75% of global pharmaceutical profits come from American taxpayers.
• Drug manufacturers benefit from generous research subsidies and enormous healthcare spending by the U.S. Government. Instead of passing that benefit through to American consumers, drug manufacturers then discount their products abroad to gain access to foreign markets and subsidize those discounts through high prices charged in America. Americans are subsidizing drug-manufacturer profits and foreign health systems, both in development and once the drugs are sold. 

 
DELIVERING ON PROMISES TO PUT AMERICAN PATIENTS FIRST: President Trump is delivering on promises for American patients that the political establishment did not believe were possible.

• On May 12, 2025, President Trump signed an Executive Order titled: “Delivering Most-Favored-Nation Prescription Drug Pricing to American Patients” directing the Administration to take numerous actions to bring American drug prices in line with those paid by similar nations.
• On July 31, 2025, President Trump sent letters to leading pharmaceutical manufacturers outlining the steps they must take to bring down the prices of prescription drugs in the United States to match the lowest price offered in other developed nations 
• President Trump has been relentless in his effort to address the unfair and outrageous prices Americans pay for prescription drugs:

o   President Trump: “In case after case, our citizens pay massively higher prices than other nations pay for the same exact pill, from the same factory, effectively subsidizing socialism aboard [abroad] with skyrocketing prices at home. So we would spend tremendous amounts of money in order to provide inexpensive drugs to another country. And when I say the price is different, you can see some examples where the price is beyond anything — four times, five times different.”

AI Patent Claims are 101 Patent Eligible Subject Matter

In the matter of ex parte Desjardins et al., the Appeals Review Panel vacates the Patent Trial and Appeals Board decision that claims concerning machine learning are not patent eligible subject matter.  The Appeals Review Panel included Under Secretary of Intellectual Property John Squires who is also Director of the United States Patent and Trademark Office.  The Appeals Review Panel criticized the PTAB for not adequately considering the teachings of the Federal Circuit’s Microsoft v. Enfish decision and stated:

Under a charitable view, the overbroad reasoning of the original panel below is perhaps understandable given the confusing nature of existing § 101 jurisprudence, but troubling, because this case highlights what is at stake. Categorically excluding AI innovations from patent protection in the United States jeopardizes America's leadership in this critical emerging technology. Yet, under the panel's reasoning, many AI innovations are potentially unpatentable-even if they are adequately described and nonobvious-because the panel essentially equated any machine learning with an unpatentable "algorithm" and the remaining additional elements as "generic computer components," without adequate explanation. Dec. 24. Examiners and panels should not evaluate claims at such a high level of generality.

The reasoning of the Appeals Review Panel is based upon the USPTO approach to patent eligible subject matter analysis. 

Anthropic Settles Copyright Suit

Wired Magazine reports that the class action copyright suit involving Anthropic in the Northern District of California (Bartz v. Anthropic) has been settled (for the most part) for around $1.5 billion US.  The article is available, here.  

The practical implications of the settlement on positioning in other copyright suits involving generative artificial intelligence remain to be seen.  In June, Judge Alsup issued an order stating, in part:

. . . This order grants summary judgment for Anthropic that the training use was a fair use. And, it grants that the print-to-digital format change was a fair use for a different reason. But it denies summary judgment for Anthropic that the pirated library copies must be treated as training copies.

We will have a trial on the pirated copies used to create Anthropic's central library and the resulting damages, actual or statutory (including for willfulness). That Anthropic later bought a copy of a book it earlier stole off the internet will not absolve it of liability for the theft but it may affect the extent of statutory damages. Nothing is foreclosed as to any other copies flowing from library copies for uses other than for training LLMs.

Decluttering the US Trademark Register from Fraudulent Filings

The United States Trademark Office has terminated over 52,000 trademark pending registrations and registrations based on widespread fraud by a foreign-filing entity.  Notably, the sanctions order states the following administrative sanctions:

(1) Permanently preclude Respondents from submitting trademark-related documents on behalf of Respondents or others; (2) Strike or otherwise give no weight to all trademark-related documents submitted by Respondents; (3) Deactivate any USPTO accounts in which contact information related to Respondents appears, and take action to prevent Respondents from creating or activating further accounts; (4) Block future financial transactions from credit cards used to pay filing fees associated with the improper submissions and/or associated with Respondents; (5) Terminate all pending proceedings identified by Serial Numbers in Exhibit A because they contain submissions filed by Respondents. . .; (6) For trademark proceedings later found to involve Respondents or containing submissions filed by Respondents, continue to strike documents, remove information, deactivate accounts, block financial transactions, and terminate proceedings.

The Press Release states:

On August 6, 2025, the U.S. Patent and Trademark Office (USPTO) issued sanctions against a foreign filing firm and terminated over 52,000 applications and registrations connected to the firm. The firm submitted documents on behalf of others without the proper authority or qualifications. To conceal this, they: Sought out the cooperation of U.S.-licensed attorneys and then misused their credentials and faked their electronic signatures; Repeatedly signed documents using other people’s names; Submitted fake specimens of use; and Misused USPTO.gov accounts. By removing these records from the trademark register, the USPTO is helping ensure the register accurately reflects trademarks that are actually being used in commerce.

Report Shows Growth in Investment in Intangible Assets

An article titled, Investment in Intangible Assets Surges, led by Funding for Software and Databases Amid AI Boom, which discusses a report by WIPO and Italy’s Luiss Business School, states, “that in the last year alone, intangible investment across 27 high-and middle-income economies grew by about 3 percent in real terms, reaching USD 7.6 trillion in 2024, up from USD 7.4 trillion in 2023.”  The article additionally states that, “investment in intangible assets has . . . increase[ed] at a compound annual rate of about 4 percent between 2008 and 2024, far outpacing tangible investment growth of just 1 percent.”  The article can be found, here.