This article offers a brief overview on patent law and the requirements inventors must meet in order to obtain a patent. In addition, I interviewed patent attorney Yudong Kim in order to get a better idea of how patent law actually operates inside the science/technology industry. In our conversation, he explains the day-to-day activities of a patent lawyer, the importance of intellectual property, and the growing challenges posed by artificial intelligence in the field. 

Yudong Kim is a patent attorney and engineer who focuses on IP protection for clients with complex technology and processes. He has particular experience in the semiconductor technology industry, having worked for over two decades at Intel Corporation as a process R&D engineer and manager.

Patent Law Overview

Patent law stands as one of the fastest-evolving fields within the legal landscape, sitting at the intersection of innovation, business, and regulation. Rooted in the U.S. Constitution—specifically Article I, Section 8, Clause 8—the patent system was established to encourage scientific and technological advancement by granting inventors exclusive rights to their discoveries. The Act also encourages them to pour time and resources into the development of new ideas, given that patent-holders are protected against scientific liability. This degree of market-protection involves exclusive control over the production, manufacturing, and sale of their inventions for a specified time, usually twenty years. In exchange for protection, the U.S Patent and Trademark Office, or USPTO—established under 35 U.S.C 1b as the arbiter of patent applications and disputes—requires immediate disclosure of patented information. After the protection period ends, patented discoveries enter the public domain [1]. 

Qualifying for patent protection involves meeting five core legal criteria: patentable subject matter, utility, novelty, non-obviousness, and enablement. Each requirement helps ensure that only inventions that are truly new, useful, and distinct from existing knowledge receive protection.

Patent Requirements:

Patentable subject matter concerns the types of inventions that are usually considered for patents. 35 U.S.C § 101 broadly defines such ‘subject matter’ as “any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor” [2]. 

Utility merely requires an invention to be ‘useful’ in the conventional sense of the term, and is split into three sub-requirements: credibility, specificity, and of a substantial nature. Credible utility means that an invention’s logic and reasoning must be adequate enough for individuals within that invention’s field to verify if it can be used as intended. Specific utility means that the invention must be subject-matter specific, while substantial utility requires an invention to have a concrete, real-world usage.   

The novelty of an invention depends on the quality of being new, and statutory bars. Being new simply means that the invention isn’t made or used in any country nor described in any printed material at the time of the patent application. Statutory bars refers to the fact that the invention must not have been publicly used, sold, or described more than one year before the patent application date. Essentially, novelty is assessed based on the state of knowledge before the invention date, while statutory bars are determined by activities before the patent application filing date. A lack of novelty means the invention cannot be patented, while a statutory bar can prevent a patent even if the invention is novel. An example clarifies this distinction: if an inventor creates a technology that no one else has ever made or described in similar terms, the technology is considered novel. If that inventor publicly uses this technology for more than a year before filing a patent application, a statutory bar will prevent the inventor from obtaining a patent, even if the technology is novel.

In order for an invention to be non-obvious, its subject matter must not be an obvious derivation of ‘prior art’—existing knowledge in a given field—and is not something a person with ordinary skill in such a field would have easily or predictably come up with. This condition was added to the patent-determining process via the Patent Act of 1952, and was applied many times by the Supreme Court. For instance, in KSR International Co. v. Teleflex, the Supreme Court rejected the Court of Appeals for the Federal Circuit’s decision that a patent claim is obvious if “some motivation or suggestion to combine the prior art teachings can be found in the prior art, the nature of the problem, or the knowledge of a person having ordinary skill in the art.” Instead, the Supreme Court held that “a patent claim is only proved obvious if the prior art, the problem’s nature, or the knowledge of a person having ordinary skill in the art reveals some motivation or suggestion to combine the prior art teachings” [3]. 

The final requirement an invention has to meet is enablement, which has three sub-requirements: the enablement requirement, written description requirement, and best mode requirement. An enabling invention must include a specification that describes the invention in sufficient detail for individuals within that invention’s art to reproduce it, as well as one or more claims at the end of the specification that precisely states the invention’s legal definition. The invention application must also include a written portion, where the inventor thoroughly describes the relationship between the specification and claims; this is done to prevent inventors from claiming inventions they haven’t yet conceived or developed. For the best mode requirement, the applicant must disclose their preferred method of practicing the invention if it materially impacts its use.

Interview:

Q: What made you want to be a patent attorney? 

Yudong: So I used to be an engineer myself. I worked at another semiconductor company, Intel, for about 20-plus years. And while I was an engineer, I was involved in the invention disclosure and patent processes. Because Intel was a leading technology company, naturally, there were a lot of legal actions against us, with which I have some experience—in terms of protecting Intel. Based on those two fronts, I got a taste of patent law practice: one being securing a company’s own patents to protect IP—intellectual property—and the second being figuring out how IP rights often play out.

For example, this is a bit of an old story, but about a year ago, Apple’s Apple Watch had the capability to measure something related to blood sugar levels. Apple was found to be infringing someone else’s patent on that feature. So Apple had to pull the Apple Watch from the market to disable that function, and evidently, the newer version of the Apple Watch doesn’t have that capability. That’s how strict—and how significant—patent rights can impact a business model.

And for Intel, about three years ago, VLSI Technology sued Intel for patent infringement on two chip-making patents, and a federal trial court actually awarded VLSI Technology two billion dollars in damages. However, Intel later challenged the verdict at the Federal Circuit Court of Appeals, and we were able to bring down the amount of damages quite a bit. So, these two examples are good examples for anyone interested in patent law practice. Oh, yeah, and litigation, like in the Apple and Intel cases, is sort of an extreme scenario. The majority of patent law practice wraps up before litigation, and during that process, there can be a lot of licensing negotiations.

So, all this was kind of interesting to me because, at that point in time, I had done 15-plus years of technology development. It’s sort of—I think once you get past 10 years or so in technology development—everything kind of repeats itself. So, I suppose I was looking for something different, and I was involved in creating my own patents. As an engineer at Intel, I was supporting those litigation activities in terms of protecting Intel’s position from a technical standpoint. That kind of exposure to the practice made me interested in patent law, and I just made a call to shift my career into patent law.

Q: Is it a crucial part of patent filing for, say, an invention, where you have to disclose your invention and write out all those engineering specifications?

Yudong: Yeah, so typically in a corporate environment, an engineer might have an idea, and then the idea will be put into a presentation and go through an internal evaluation process. If the idea is deemed patentable, it takes about $50,000 to write a patent, apply for it, and receive it. So although this depends on the company culture, most companies are reasonably alert about protecting their own inventions. But there’s also the fact that companies have limited resources to pour into patents.

So let’s say there are 10 ideas. After the engineers who came up with those ideas put them into some kind of communicative writing, I would say half—or slightly less than half—get selected to actually go through the patent process. What follows is very company-dependent, but typically, technical companies have some kind of disclosure form that needs to be filled out to describe the invention. Then there’s a review process. Once an idea is selected to go through the patent application process, patent attorneys are assigned to the case for that disclosure, and they hold disclosure meetings with company heads and engineers. Afterward, those patent attorneys take the ball and write the patent to describe the invention, and then finish the patent application process.

So basically, the invention disclosure aspect of patent law is related to engineers putting their ideas into a certain document or writing, and then going through an evaluation process. During the evaluation, engineers may be called upon to clarify certain parts of their ideas—it’s all a very internal process.

Now, let’s say after an engineer comes up with an idea and goes through that internal disclosure process, the engineer wants to publish a paper on the idea he or she is patenting. It’s a very cool idea, so they may publish the process, the invention, or something describing the invention. Then, their patent rights might actually become foreclosed because the engineer disclosed it to the outside world. And actually, I didn’t know about this when I was an engineer, so I published a lot of papers, but looking back, I’m thinking now I shouldn’t have. So those kinds of things are what’s involved with invention disclosures.

Q: How do you see yourself incorporating the knowledge that you learned in your education on these patents/patent suits?

Yudong: At university, when you’re dealing with clients or reading all this news about patent infringement, you realize that the legal requirement for a patent—of a patentable idea—is that it has to be something new and non-obvious. If it’s something new but obvious to put together from known facts, then it’s not patentable.

I would say what I learned from school—like graduate school in those days—doesn’t really apply in terms of securing your own patent. Those are fundamental ideas upon which I can build my own ideas, so getting a good education is a very crucial part of it. But in and of itself, it doesn’t provide any opportunity to pursue a patent on it. Because it’s already known, right?

But in the research environment, for example, certain advisors—like graduate school advisors, professors, and so forth—are often more aware of patent aspects, so they try to acquire patents. But then their roadblock is the cost. To write up the patent and go through the patent-obtaining process and so on and so forth, it’s costly, especially in academia. So that’s a little bit of a roadblock.

But of course, certain professors are aggressive in terms of acquiring patent rights and IP rights on their research subjects, and along those lines, PhD students under them who assist in their research projects may actually become co-inventors of the patent. So yeah, I would say at the undergraduate level or master’s degree level, there may be a limited chance of getting a patent of your own. But at the PhD level, there can be opportunities for getting some exposure to the patent process by being a co-inventor of some research results or ideas.

Q: How do you stay current with advancements in technology and patent law, especially given the fast pace of innovation these days?  

Yudong: In terms of patent law, it’s one of those areas within the legal field where there can be somewhat frequent changes. For example, First Amendment rights might not change over multiple decades, but in patent law, changes can happen within five or six years. However, from the perspective of technology transition or advancement, the pace is relatively slow compared to the speed of innovation itself. So, I don’t worry too much about patent law evolving into something drastically different. Instead, I focus on staying informed about what’s changing and what’s on the horizon. That said, keeping up with changes doesn’t take up significant bandwidth for me, which is a relief.

On the other hand, when it comes to technology itself, I do need to stay updated on advancements. It depends on the area of technology. For example, artificial intelligence (AI) is a vastly changing field. However, in terms of patents, activity in AI has been ongoing since the late 1970s and 1980s, so some fundamentals are already well-established. Still, as a patent attorney, it’s important to stay informed about ongoing changes, which can be challenging.

For instance, I work in the semiconductor technology area. When I decided to pursue a career in patent law, I spent about six years at law firms because they have expertise in patent law practice. I wanted to gain hands-on experience dealing with disclosures, working with inventors, and drafting patent applications. After that, I transitioned to a corporate environment at Texas Instruments (TI). Being an in-house attorney, I have closer interactions with inventors because I work directly with engineers. I can walk down the aisle and talk to them about technical questions, advancements, or trends.

In-house patent attorneys are relatively small in number compared to the total population of patent attorneys, but I chose this path because of the close interaction with engineers, who are at the forefront of technology. From a technological perspective, patent attorneys must dedicate time and effort to understanding and keeping up with advancements.

Q: How do you approach patent prosecution for AI and other advanced, emerging technologies, especially when there may be less precedent in place?  

Yudong: AI is a fascinating area, but one of the most significant challenges is that it often involves mental processes, which are not inherently patentable under the law. Patentable subject matter is defined in the Constitution, and mental processes alone do not qualify. The challenge, then, is finding ways to tie the mental process to some hardware or a routine performed by a machine or computer.

Patent examiners at the USPTO (U.S. Patent and Trademark Office) ensure that newly issued patents are novel and non-obvious. They rigorously assess whether an idea is already disclosed by prior art, such as existing patents or publications, or whether it’s an obvious combination of known technologies. So, in the AI domain, it’s critical to demonstrate that the claimed invention involves more than just mental processes—it must require machine involvement.

Q: How do you ensure that a patent claim is broad enough to provide adequate protection for an invention, but also specific enough to overcome challenges from prior art?  

Yudong: I rely heavily on inventors for this. They are the subject matter experts and can best describe what’s novel about their invention compared to what already exists. My first step is to understand this distinction clearly during the interview process. From there, I translate the inventor’s description into claim language, which can be challenging for engineers to interpret initially. However, once you get the hang of it, it’s essentially English writing with a specific format.

Typically, a patent claim includes known elements to provide context, in addition to the novel aspects that make the invention unique and non-obvious. For example, a transistor has been known to have three terminals—collector, emitter, and base—since the 1960s. But if the base is configured differently in a novel way, that distinction must be included in the claim. Ultimately, the new aspects of an invention, as described by the inventor, form the basis for what’s claimed.

Q: Is there overlap between patent law and corporate law?  

Yudong: Patent law and corporate law are very distinct. Patent law focuses on science and technology at a high level, while corporate law deals with aspects of business operations, such as compliance with government regulations, corporate ethics, and human resources. Corporate law might also include mergers and acquisitions, but these are separate from patent law practices. For instance, if Company A acquires Company B, Company B’s patents become the property of Company A after the merger. However, the patents themselves remain intact and are unaffected by the merger. Patents are intellectual property with a 20-year lifespan, after which their subject matter becomes public knowledge. 

Q: Is there any common law or like judicial precedent you follow when you’re dealing with patents?  

Yudong: Not really. Patent law is based on the U. S. Constitution; so it’s a federal law, not a state law.  In terms of the Supreme Court interpreting patent law, there’s usually significant precedent, but I wouldn’t say it’s like common law-based like contract law or torts law, which are more state laws based on common ground. 

*This interview has been edited for clarity.