‘Joint Research’ Doesn’t Guarantee Joint Ownership: IP Pitfalls in U.S.-Korea Technology Collaboration

Technology collaboration between the United States and Korea has grown quickly, from semiconductors and artificial intelligence to biotechnology and advanced manufacturing. More joint research agreements are being signed, but the importance of their IP provisions is often overlooked. This can cause problems later, because the two countries have different legal rules that even experienced counsel can miss.

Two points are especially important. First, paying for a project does not by itself give ownership of the results. Second, a co-owner’s rights differ greatly depending on whether the patent is a U.S. patent or a Korean patent. This article covers the basics of contract drafting and research scope, ownership, the use of jointly owned patents, and other practical issues to address before signing.

1. The Basics: Contract Drafting Practices, Research Scope, and Background IP

Before turning to specific IP issues, one point needs to be addressed. U.S. and Korean lawyers approach contracts differently. U.S. contracts follow the common-law tradition and are long and detailed, because U.S. drafters try to address every situation in the text itself. Korean contracts, by contrast, follow the civil-law tradition and are often much shorter: the parties agree on the core terms, such as scope, ownership, term, and key obligations, and statutes such as the Civil Act and the Commercial Act supply the rest as default rules.

Returning to the joint research itself, the most basic element is the research scope. It matters for patent filing and prior art, not only for ownership. Under 35 U.S.C. § 100(h), a “joint research agreement” is a written contract, grant, or cooperative agreement for research work “in the field of the claimed invention.” Under Section 102(c), subject matter disclosed by one party and an invention claimed by another are treated as commonly owned, so the first party’s earlier patent filings cannot be cited as prior art against the second party under Section 102(a)(2). This applies only if the agreement was in effect by the effective filing date, the claimed invention resulted from activities within the agreement’s scope, and the application names the parties. The meaning of these two provisions is clear. Calling a project “joint research” is not enough. The agreement should describe the research field, participants, and handling of results clearly enough to match any later patent application.

Once the scope is defined, the parties should separate Background IP from Foreground IP. Background IP is what each party already had before the project, or developed outside its scope. It includes not only patents but also trade secrets, experimental data, materials, and source code. Foreground IP is what the joint research newly creates, and only Foreground IP should be the subject of joint filing and joint ownership. If one party’s existing assets become joint IP simply because they were used in the project, that is not fair collaboration. A schedule listing each party’s Background IP, and the research-only license terms under which it is shared, prevents most of these disputes.

Additionally, one difference between the two legal systems deserves special attention: Korean contract law has no concept that corresponds to consideration. A Korean contract becomes binding through the agreement of the parties alone. A short Korean-style research agreement is therefore not incomplete just because it lacks recitals or the exchange of promises a U.S. lawyer expects to see. At the same time, a Korean partner may see a long U.S. draft as overly complex, or even as a sign of distrust. It helps to agree on the governing law early, and to make clear which issues the contract will decide and which will be left to default rules.

2. Ownership: Two Things to Check

With the research scope and Background IP settled, the next question is who owns the results. The answer depends on two things: the basic allocation rule, and any conditions that come with government funding.

For cooperative activities between the two governments, the basic rule comes from the U.S.-Korea Agreement Relating to Scientific and Technical Cooperation (the “S&T Agreement”). Under Annex I, Section III.B(2)(a), IP created by persons employed or sponsored by one party belongs to that party, and IP created by persons employed or sponsored by both parties is jointly owned. Private agreements are not bound by this annex, but many follow the same logic.

The key point is that ownership depends on who created the invention and on their employment or sponsorship relationship, not simply on who paid for the project. A U.S. company that funds a project led by a Korean university does not obtain rights to the results just by paying, and many disputes come from this gap between business expectations and the legal rule. Some “joint research” projects are also, in substance, technology transfers, where the IP stays with one party unless sharing is separately negotiated. The agreement should therefore identify the participating researchers from each side and confirm that their employment and sponsorship relationships match the intended ownership structure.

Government funding adds another layer. In Korea, Article 16 of the National R&D Innovation Act and Article 32 of its Enforcement Decree provide that, when research is conducted jointly with a foreign institution or individual, the results should belong to the Korean institution, or the Korean institution should receive priority to use them. In the United States, the Bayh-Dole Act generally allows nonprofit organizations and small businesses to keep title to federally funded inventions. However, under 35 U.S.C. § 202(a), a funding agreement may provide otherwise when the contractor is not located in the United States, has no U.S. place of business, or is subject to the control of a foreign government. In practice, a foreign institution may not be allowed to elect title at all.

Both systems, in short, expect government-funded results to be used domestically. Because of differences in bargaining power, full joint ownership is not always possible. In that case, securing a license, a right of first negotiation, or a right of first refusal to buy the other party’s share can still be regarded as faithfully complying with these laws.

3. Using Jointly Owned Patents: Where U.S. and Korean Law Differ the Most

Even after ownership is settled, a harder question often remains: what can each co-owner actually do with the patent? Using a patent includes licensing and assigning a share, not only making products. On this point, U.S. and Korean law take opposite approaches.

Under 35 U.S.C. § 262, each co-owner of a U.S. patent may use the invention without the other owners’ consent and without accounting to them, and U.S. courts allow each co-owner to grant non-exclusive licenses on its own. Meanwhile, under Article 99 of the Korean Patent Act, each co-owner of a Korean patent may practice the invention itself, but it needs the consent of all other co-owners to assign its share or to license a third party.

These rules depend on where the patent is registered, not on the nationality of the owners. Joint research often produces patents in both countries, so the same partners may face opposite rules for the same invention. For example, a Korean company may assume that its U.S. partner cannot license their jointly owned patent without its approval. Yet the U.S. partner can license the U.S. patent to anyone, even the Korean company’s competitor, without asking. In the opposite case, a U.S. company may assume that it can freely license the Korean patent, only to learn that it needs its Korean partner’s consent.

The agreement should therefore not rely on default rules. It should state whether third-party licensing and assignment are allowed, how consent will be obtained, and whether transfers to competitors are restricted. A right of first refusal, which lets the other co-owner match any third-party offer, is another useful tool. Joint ownership is only the starting point; what each party can actually do depends on the agreement.

4. Other Practical Issues to Address Before Signing

Several other issues are easy to miss in a first draft and costly to fix later.

Foreign filings. The agreement should decide who is responsible for filing, and paying for, applications outside Korea and the United States. Universities often want protection only in their home country, while companies usually want wider coverage. A desirable approach is to let the party that wants protection in a third country pay the costs and hold the rights there, while the other party receives a license to practice the invention in that third country and agrees to cooperate with the filing. Two legal limits also apply. First, an invention made in the United States may not be filed abroad without a foreign filing license from the U.S. Patent and Trademark Office (USPTO). In practice, most applicants file in the United States first, and the license is usually granted with the filing receipt. Second, both countries have secrecy order and export control systems for sensitive technologies, which can restrict foreign filing and the disclosure of technical information.

Trade secrets. Joint research produces much information that will never be patented, such as experimental data, know-how, and source code. Such information often has value only while it stays secret, so protecting it as a trade secret is important. Marking a document “confidential” is not enough in either country; the owner must show that it actually managed the information as a secret. An NDA or strong confidentiality clauses, a list of protected information, access controls, and duties to return or destroy materials all help.

Publications and presentations. Research results are often made public through conference presentations, exhibitions, and demonstrations, not only through papers. Both countries offer a 12-month grace period, but the conditions and procedures differ, so the parties should check the disclosure rules of each country where they plan to file. The agreement should also require advance notice, for example written notice and the other party’s consent 30 days before any publication or presentation to prevent unexpected or unintended disclosure. At the same time, one party should not be able to delay publication without limit. Under Annex I of the S&T Agreement, a party may request a delay to protect its rights in an invention, but unless otherwise agreed in writing, the delay may not exceed six months. This is a useful reference point for private agreements as well.

Non-infringement warranties. A warranty that jointly developed technology does not infringe any third-party rights is risky, because no party can be fully sure of this. It is safer to limit the warranty to what a party knew or reasonably should have known, and to add a cap on total liability. A dispute over a poorly drafted warranty can cost more than the research itself.

Resolve Issues Early

IP provisions in U.S.-Korea joint research agreements need more than standard-form treatment. The first steps are to understand each other’s contract drafting practices, define the research scope, and separate Background IP from Foreground IP. Ownership follows the researchers rather than simply the funding, and government funding adds conditions on both sides. The different co-ownership rules, as well as foreign filings, confidentiality, publication, and warranties, should be addressed in the first draft. These issues are easy to resolve at the drafting stage, but much more expensive to resolve after a dispute arises.

Image Source: Deposit Photos
Author: gannenko
Image ID: 689813496