By Kirsten Potts
A ten-minute train ride outside of Geneva leads to one of the most significant scientific collaborations in the world: the European Organization for Nuclear Research (CERN). Established in the 1940s, CERN brings together scientists from twenty-five member states and several observer countries, including the United States.[1] At the core of the collaboration lies a deep institutional commitment to open science. It has further cemented this commitment by its release of its new Open Science Report, designed to evaluate how well it has been adhering to its open science commitments. Yet CERN’s open science policy creates a persistent tension with patent law and the commercialization of technology. This tension contrasts sharply with the prevailing approach in the United States, where those who believe a discovery should be patented frequently prevail over those who think it should be placed in the public domain.[2] The United States has actually institutionalized patenting government-funded research through the Bayh-Dole Act, which was designed to promote widespread utilization of federally funded research.[3] CERN’s institutional commitment to openness challenges this presumption by prioritizing broad access to scientific knowledge over exclusivity, even when discoveries may have commercial potential.
According to UNESCO, open science seeks to make scientific knowledge accessible to all, so that society might benefit from research beyond the scientific community itself.[4] Consistent with this vision, CERN’s Open Science Policy commits the organization to open access to publications, open data, open-source software, open hardware, research integrity, reuse, and reproducibility, robust research infrastructures, research assessment and evaluation, and education and outreach.[5] While these commitments promote accessibility, they also create challenges when CERN discoveries are capable of being translated into commercially viable technologies.
This tension reflects a foundational principle of intellectual property law: for many technologies, particularly those requiring substantial development costs, commercialization depends on the willingness of private actors to invest time and resources in exchange for some degree of exclusivity. While broad dissemination may be appropriate for unpatentable scientific discoveries, it is often ill-suited for innovations that exhibit commercial potential and require significant resources to develop. The biotech sector illustrates this dynamic clearly. Developing a new pharmaceutical product requires billions of dollars in investment, and without patent protection, private investors would have little incentive to commit such resources if the resulting product could immediately be copied and distributed freely. Ironically, without this investment, even a life-saving drug could remain inaccessible to the public.
At the same time, publicly funded research institutions face competing obligations: to the public and to themselves. Releasing inventions into the public domain may serve as a means of “repaying” the public for its investment.[6] CERN thus occupies a unique position, balancing public accountability against the practical realities of technology development.
CERN attempts to reconcile these competing pressures through targeted patenting and structured technology transfer. Its Knowledge Transfer Office works with scientists, engineers, and industry partners to identify innovations with potential technical or commercial applications.[7] Many CERN-developed technologies have relevance beyond high-energy physics, including applications in medical imaging, materials science, and computing. In such cases, patent protection may be necessary to attract industrial partners willing to invest in development and deployment.[8]
Although CERN patents selectively, it identifies three circumstances in which patent protection is appropriate: when patenting facilitates knowledge transfer beyond high-energy physics; when it enables further non-patentable innovation; and when CERN intends to license the technology to external entities.[9] This approach reflects an effort to preserve openness while still enabling commercialization where necessary.
Patent law’s novelty requirement, however, presents a particular challenge for CERN. Because novelty is destroyed by public disclosure, CERN’s culture of rapid publication and information sharing creates a heightened risk that patentable inventions will become unprotectable prior art. This risk is amplified by CERN’s large, rotating collaboration model. Researchers, students, and visiting scientists frequently participate for limited periods, and many are unfamiliar with the legal implications of disclosure. Even internal disclosures may jeopardize patentability if individuals present are not bound by confidentiality obligations. In such an environment, accidental prior art is a significant concern.
Moreover, CERN’s institutional culture encourages early and enthusiastic sharing of discoveries, a value central to its scientific mission but one that must occasionally be restrained to preserve patent eligibility. These constraints help explain why CERN’s open science model remains particularly attractive: openness reduces administrative friction and allows CERN to prioritize training, collaboration, and accessibility, even if some patent opportunities are foregone. Furthermore, CERN protects its investors by making its data available several years after the experiment.[10]
CERN’s status as an international organization further complicates its relationship with patent law. The organization is accountable to member states with diverse patent systems, uneven industrial capacity, and differing public-interest expectations. Because patent rights are territorial, CERN cannot obtain a single global patent and must instead choose where to file, an inherently political decision that risks privileging certain member states over others. As a result, CERN has adopted a cautious approach to patenting, reflecting institutional and political constraints rather than purely philosophical opposition to intellectual property. CERN’s governance structure – a council made up of one scientist and one administrator from every member country – has allowed it to continue functioning well.[11] It allows the organization to remain committed to its mission and committed to cooperation between the countries.[12]
Furthermore, open science models can reduce governance and transaction costs. Maxwell Morgan and E. Richard Gold argue that open approaches to drug development may succeed where institutions possess the organizational capacity to manage legal, regulatory, and funding complexities.[13] CERN’s governance structure, built around shared scientific objectives among member states, has enabled it to implement such a model more effectively than many other research institutions.
Some CERN technologies have clearly benefited from openness. The decision not to patent the World Wide Web facilitated the emergence of a unified global system rather than fragmented proprietary alternatives.[14] Yet not all CERN innovations resemble the Web. Many technologies involve high development costs and significant technical risk, making private investment, and therefore some form of exclusivity, essential.
Ultimately, CERN’s strategy reflects a deliberate tradeoff. Its open science policy has enabled extraordinary scientific collaboration, workforce training, and public access to knowledge. At the same time, this commitment inevitably constrains certain avenues of commercialization. Given its multinational governance and public mission, CERN has settled on a pragmatic middle ground, one that illuminates the broader tension between openness and intellectual property law in modern scientific research[1] .
[1] Our Member States, CERN, https://home.cern/about/who-we-are/our-governance/member-states?utm.
[2] Rebecca S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 Va. L. Rev. 1663, 1666 (1996).
[3] Arti K. Rai & Rebecca S. Eisenberg, Bayh-Dole Reform and the Progress of Biomedicine, 66-SPG Law & Contemp. Probs. 289, 290 (2003).
[4] UNESCO, UNESCO Recommendation on Open Science (2021), https://unesdoc.unesco.org/ark:/48223/pf0000379949.locale=en.
[5] CERN, CERN Open Science Policy (2022), https://cds.cern.ch/record/2835057/files/CERN-OPEN-2022-013.pdf.
[6] Rebecca S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 Va. L. Rev. 1663, 1667 (1996).
[7] CERN, Knowledge Transfer: About Us, https://knowledgetransfer.web.cern.ch/about-us.
[8] CERN Knowledge Transfer Group, CERN Policy on the Use of Patents as a Tool for Knowledge Transfer (2019) https://knowledgetransfer.web.cern.ch/system/files/resource/intellectual-property-reference/patent-policy/cern-patent-policyfvclean_0.pdf?utm.
[9] Id.
[10] Elise De Geyter, Open Access Policy in International Organizations, 2017 WL 6326473 (2017).
[11] Kevin Fraizer, The Need for an International AI Research Initiative: How to Create and Sustain a Virtuous Research-Regulation Cycle to Govern AI, 19 Wash. J. L. Tech. & Arts 62, 86 (2024).
[12] Id at 88.
[13] Maxwell Morgan & E. Richard Gold, From Patents to Regulatory Exclusives in Drug Development: A Comparative Transaction Cost Analysis, 64 Jurimetrics J. 375, 76 (2024).
[14] CERN, The Birth of the Web, https://home.cern/science/computing/birth-web.