In his essay, “Unpacking My Library,” philosopher Walter Benjamin wrote that “of all the ways of acquiring books, writing them oneself is generally regarded as the most praiseworthy method.”

The same is true – even more so – for patents.

While many firms recognize the value that patents can provide (including attracting investment, enhancing M&A value, blocking competitors and generating licensing revenue), they often assume that filing for a patent will require excessive time, cost and effort. And, particularly in the fields of tech and finance, firms are often skeptical that patents on relevant technology can be obtained or meaningfully enforced. As a result, founders in the venture, crypto, Web3 and AI space may give only cursory thought to patent protection before deciding to focus their resources elsewhere.

Decasonic, a Chicago‑based venture and digital assets firm, provides a case study that pushes back on these assumptions. By strategically prioritizing the intellectual property they wished to protect, working with the right law firm and taking advantage of recent policy changes at the US Patent Office, Decasonic was able to quickly and cost-effectively file an AI patent application.

“We were surprised by the efficiency of the process and how little time it required from us,” says Paul Hsu, CEO of Decasonic.

(Decasonic closed on $48.8 million for Decasonic Disruptive Innovation Venture & Digital Assets Fund I in January 2022, according to PEI data. LPs in the fund included venture firm Drive Capital.)

Decasonic, a team of investor operators investing at the frontier of Web3 and AI, partnered with Norton Rose Fulbright to seek patent protection for an agentic AI platform, the Reinforcement Learning Expert Network (RLEN).

RLEN is a dynamic, adaptive network of human and AI experts, where humans engage with AI clones of world-class specialists to fine-tune models, inject real-world context and sharpen the intelligence produced by the network. The clones within RLEN explore non-consensus and subjective ideas, synthesize expert signals and translate shared clone intelligence into actionable insights for early-stage venture investing. The goal is to leverage the expertise of world-class experts (as clones) to sharpen venture decisions in domains where timing, context and subject matter expertise matter most.

Through RLEN, Decasonic is capable of receiving the combined intelligence of world-class domain experts and improving itself through continuous human usage. “In areas where there is non-consensus, subjective evaluations and investment risk, RLEN delivers a faster and deeper set of applied insights that enhance our investment alpha,” says Abdul Al Ali, an investor at Decasonic.

RLEN reflects Decasonic’s broader AI vision: building systems where humans and AI continuously flourish each other. As a core layer of the firm’s internal AI Operating System, the network is designed to learn from outcomes, feedback loops and experience over time, evolving toward a reinforcement learning-driven operating system that steadily improves specialized agentic teammates across core venture workflows. In 2026, the team is focused on orchestrating the foundational stack (OS, applications, agents) with memory and context so intelligence, context and signals flow across internal and portfolio work.

The decision to pursue patent protection for RLEN was especially timely given significant pro-patent policy changes implemented at the United States Patent and Trademark Office (USPTO) over the past few months. These include guidance that provides more favorable treatment for inventors seeking patents in the software/machine learning/AI space, as well as limitations on procedures that have been used to challenge the validity of issued patents.

These changes have been driven by John Squires, who took the helm as the 60th Director of the USPTO in September 2025. Earlier in his career, Squires served as chief IP counsel at Goldman Sachs, led the creation of the US’s first patent asset-backed finance platform for one of the world’s leading funds and served as partner and chair of the IP and Emerging Companies Practice at his most recent law firm.

Just two days after being sworn in as director, Squires held a ceremonial signing to issue two patents directed to technology areas that have generally been viewed as problematic for patent eligibility – a blockchain patent in the fintech space and a cancer antibody patent in the pharma space. At that signing, Squires said that “the USPTO is open for business, especially for the technologies of tomorrow.” He later noted that “if there is any one thing I am going to accomplish in my tenure, it’s this — making sure the door to the patent office is wide open to transformative technologies.”

Just days later, Squires issued an opinion overturning a USPTO decision that found that an application directed to machine learning was not patent eligible, noting that “categorically excluding AI innovations from patent protection in the United States jeopardizes America’s leadership in this critical emerging technology.” He has been particularly focused on AI patents, noting that “without reliable patent protection, AI start-ups cannot secure the venture capital needed to compete against state-backed giants in China and elsewhere.”

Squires has also enacted policy changes that make it more difficult for alleged infringers to challenge patents. In 2012, the America Invents Act created inter partes review (IPR) proceedings, which provide a streamlined process at the USPTO for companies to challenge the validity of issued patents. On average, the USPTO had been “instituting” such reviews of challenged patents more than 65 percent of the time. That figure has dropped to approximately 17 percent since Squires became director.

Obtaining a patent is no guarantee that it can be asserted successfully against competitors, or that the courts will agree with the USPTO that the patent should have been issued. But for Decasonic, seeking patent protection was about more than just safeguarding its intellectual property. The patent serves as proof of their long-term vision to lead innovation with integrity and impact, attract founders aligned with that vision and signal a long‑term commitment to building their business with intelligence at its core.

“Going through the patent application process reinforced the importance of precision and purpose in describing how innovation creates value,” Al Ali explains. “Patents are a tool for protection, but they also serve as a way to articulate a firm’s contribution to its industry. This has made us more intentional in defining how our internal AI development creates impactful value-add for portfolio company founders, investors and partners.”

John McBride is a partner in the Chicago office of law firm Norton Rose Fulbright. He is a trial lawyer who represents technology companies and other clients in patent and trade secret cases. He may be reached at John.Mcbride@Nortonrosefulbright.com. McBride co-authored this column with Paul Hsu, founder and CEO of Decasonic, and Abdul Al Ali, an investor in Decasonic.