I. Introduction
In a 1955 interview with Jonas Salk, the inventor of the first polio vaccine, when asked whether he owned the patents for the vaccine, Salk replied with a rhetorical question: “Could you patent the sun?” While Salk chose not to patent the polio vaccine, believing it should be universally available due to its critical importance to public health, this choice was unusual. This is because patents have been the key regulatory instrument for commodifying knowledge so that it has exchange value in the market. On that basis, inventors or patent owners can profit from the invention.Footnote 1 Patents have also been identified as a key barrier to the equitable distribution of vaccines during pandemics, including COVID-19.Footnote 2 Ironically, seven decades after Salk’s interview, patents have been applied for and granted to technologies directly related to the Sun.Footnote 3 Specifically, these refer to the solar radiation modification (SRM) technologies, which are designed to cool the Earth by introducing tiny reflective particles into the upper atmosphere, enhancing reflective cloud cover in the lower atmosphere, or reducing the thickness of high-altitude clouds that can absorb heat.Footnote 4
The intellectual property (IP) system, particularly the patent system, is often said to aim to provide incentives for inventors and creators by granting them exclusive rights for a limited period. Over time, the IP system has been justified by various philosophical foundations, including utilitarianism, natural rights and labour theory, and IP instrumentalism.Footnote 5 The institution of IP has been identified as a barrier to international clean technology diffusion.Footnote 6 However, amid international legal fragmentation in sustainability governance and IP regulation, the Sustainable Development Goals (SDGs) are neither embedded nor prioritised in international IP law.Footnote 7 The trade-climate nexus primarily follows a liberal environmental approach, making it challenging to promote trade-conditioning measures within already commercially biased policy subsystems.Footnote 8 The prolonged international negotiations on the diffusion of clean technology also make it difficult to reach an agreement on a blanket restriction on clean technology patents. Given the substantial differences among technologies, answers to both questions will be fact-specific, technology-specific, market-specific, timing-specific and instrument-specific. There seems to be no one-size-fits-all solution. This calls for specific case studies to understand the nature of the technology, the potential benefits and harms it may bring, the interests it may impact, and the limits of current international regulatory systems. This article contributes to the literature on the intersection of IP and climate change through a case study on patenting and SRM technologies.
Unlike climate mitigation technologies that have clear positive climate impacts, SRM has inherent dual effects – its potential benefits in helping reduce climate risk, limit suffering, and lessen ecosystem degradation coexist with significant potential risks, including environmental harm and governance challenges.Footnote 9 The development of SRM has attracted sharply increased attention and controversy, with proposals to expand research and governance consultations alongside proposals to restrict or prohibit its research and deployment.Footnote 10 Among the regulatory mechanisms for SRM governance, the IP system, particularly the patent system, may play an important role.Footnote 11 Due to concerns about risks, misuse, or overreliance, there have been proposals to govern IP rights for SRM technologies within the existing frameworks of international and national IP laws, including restrictions on granted patents (i.e., ex post restrictions; see Section II.2).
However, recent international developments have begun to put patenting for SRM in the spotlight. On the one hand, while private actors, prominently from the USA, have been actively seeking patents for their technologies,Footnote 12 the 2023 White House Office of Science and Technology’s (OSTP) reportFootnote 13 was considered the first major government-led research plan for SRM technologies globally. It has indicated strong support for expanded research, including outdoor experimentation, scenario-based research and monitoring capabilities to detect potential deployment.Footnote 14 It does not mention IP or patents, but that does not diminish the importance of IP. The silence suggests a business-as-usual stance – existing IP law is sufficient and no restrictions on SRM patenting are needed.Footnote 15 On the other hand, a group of scientists have advocated for a global commitment to refrain from developing or deploying these technologies via an International Non-Use Agreement on Solar Geoengineering, calling for prohibiting patent rights for SRM technologies, including technologies such as retrofitting aeroplanes for aerosol injections.Footnote 16 The EU Science Advice for Policy by European Academies (SAPEA) report also leans toward responsible research and away from relying on SRM as a response to the climate crisis.Footnote 17 Regarding IP, it was recommended that “the EU could consider regulating or prohibiting patenting and IP protection for any potential technology development relating to SRM, with a view to preventing undue accumulation of IP in private hands or other sources of pursuit of SRM not driven by the public good.”Footnote 18 This more cautious and responsible approach echoes the concerns from the Global South, evidenced by the African group’s opposition to the Swiss proposal to relax the de facto moratorium on SRM deployment at the 2024 Nairobi UN Environmental Assembly.Footnote 19
The recent prohibition proposals provide a useful starting point for considering how to design context-specific and technology-specific grounds for prohibiting or restricting patents for particular SRM inventions. The true question, however, is how such a prohibition could be implemented at all under the current international IP law. It was argued that filing a patent does not authorise their deployment if they are banned by law, and so the issue of patents is relatively secondary in the context of a Non-Use Agreement.Footnote 20 However, the history of global governance in the climate-trade nexus suggests the opposite.Footnote 21 The WTO Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) still plays a dominant role at the intersection of international regulation of IP and sustainability (which includes international environmental law, international human rights law and any proposed sui generis regulations). The TRIPS Agreement itself does not have the capacity to effectively exclude patenting for certain SRM technologies by incorporating alternative normative objectives. Specifically, under the TRIPS Agreement, due to its principle of technology neutrality (Article 27.1), no special treatment is allowed for SRM technologies. Therefore, any proposal beyond the original TRIPS mechanisms of limitations and exceptions to patents needs to be carefully designed to avoid violating the TRIPS Agreement.
The existing literature acknowledges that IP remains a nascent or contested issue in the sociotechnical system for SRM.Footnote 22 However, it primarily discusses IP and SRM technologies from an IP law perspective, taking the TRIPS principle of technology neutrality as a default position. The literature does not sufficiently consider the challenges of international coordination or how IP law might respond to scientific evidence or other legal developments. This article proposes to address the following two significant gaps in the literature – first, how undue accumulation of SRM patents in private hands or other sources not driven by the public good may pose pressure on leaning towards SRM deployment and lead to significant inequity if deployed; second, how existing international IP law, in particular the TRIPs Agreement makes it challenging to prohibit patenting for SRM technologies. Given the nature and unknowability of potential deployment of SRM technologies, this article further draws on the literature on risk regulation, which may necessitate prohibiting SRM patenting on risk-based grounds. Specifically, this article proposes embedding ex ante-based risk regulation into patent law and creating a sui generis regulatory system that includes a patent prohibition system for SRM technologies. In doing so, it contributes to the literature by connecting the isolated areas of IP law and risk regulation regarding SRM, with a solution that responds to the broader risk concerns, enables more equitable power distribution, and better resonates with the interests of the climate-vulnerable communities, rather than just protecting the commercial interests of patent right holders.
The rest of the article proceeds as follows. Section II discusses the relationship between patents and SRM technologies, including the current global patent landscape for SRM, potential grounds for excluding SRM patents, and the challenges posed by the TRIPs Agreement. Section III seeks to overcome the challenges by framing risk regulation as a patent-related issue and further provides grounds to prohibit patenting from a risk-regulation perspective. On this foundation, it proposes a sui generis regulatory system for SRM technologies that includes risk regulation and the prohibition of certain SRM patents.
II. Patents and SRM technologies: landscape, proposals for prohibiting SRM patents and the TRIPS principle of technology neutrality
1. Current patenting landscape for SRM
a. The role of patents in SRM governance
Before examining the patenting landscape of the SRM, the role of patents in SRM regulation will be briefly outlined. Patents create a monopoly upstream for a specific technical solution, excluding any third party from “making, using, offering for sale, selling, or importing” the subject matter without the patent owner’s consent.Footnote 23 The patent system, in its current design, is set on the foundation of proprietarism, which says “the possessor should take all, that ownership privileges should trump community interests and that the world and its contents are open to ownership.”Footnote 24 For the same subject matter, a property right will be granted to the applicant who first files the patent application. Even if other teams independently develop the same technical solution, only one patent right will be granted to the first filer. There is also no embedded benefit-sharing within the patent system, even when a patented invention is developed using other resources (such as genetic resources and/or associated traditional knowledge).Footnote 25
Human inventions are cumulative. However, granting broad building-block patents at the early stage of technology development can make subsequent innovation and the diffusion of relevant knowledge more costly.Footnote 26 The patent system may also hinder collaboration between scientists, as co-ownership of patent arrangements may never be satisfactory.Footnote 27 Notably, Upstream patent rights, initially offered to attract additional private investment, are increasingly viewed as entitlements by those conducting research with public funds. The UK-based Stratospheric Particle Injection for Climate Engineering (SPICE) project cancelled its tests due to the conflict of interest over co-ownership of patent applications.Footnote 28
The SPICE project indicates the role of patents in delineating interests between funders and scientists. More importantly, it is also a key instrument in regulatory capitalism, delineating private and public interests. Specifically, it is the process of commodification of knowledge – all knowledge has use value, but only patented knowledge has exchange value in the market. Once patented, knowledge is behind a paywall at a monopolistic price.
After a patent is granted, the price charged is always higher than the marginal price, as the owner is the sole supplier of the technical solution. This will lead to an undersupply and unreasonably high prices for access to the technology, as well as an equity issue regarding who can access the limited supply, as evidenced repeatedly by the problems of access to medicine and vaccines during pandemics.Footnote 29
The upstream SRM patents may involve the patent owner in SRM governance issues – if deployment is viable, they have monopolistic control over the technology as the only supplier; if deployment is prohibited, their interests are affected. Most likely, they will argue for deployment based on Article 27.1 of the TRIPS Agreement that “patent rights [shall be] enjoyable without discrimination as to … the field of technology.” Further, the free driver problem, where a single actor (state or non-state, commercial or philanthropic) may initiate deployment without multilateral authorisation and outside of international oversight, can exacerbate the patent monopoly if deployment is permitted. In addition to concerns about “rogue” actors, patent is the key to the commercialised dynamics driving SRM development.Footnote 30
In the case of SRM, relevant knowledge will have a planetary impact and should be provided as global public goods.Footnote 31 Therefore, there is a strong argument for relevant knowledge to remain an intellectual commonsFootnote 32 rather than privately owned IP. This will be elaborated in Section II.2.c.
b. SRM funding and patenting landscape
A 2014 study identified twenty-eight patent applications directly related to SRM technologies, linked to 910 family members.Footnote 33 Regarding the profile of applicants, it is clear that relevant research is dominated by public funding organisations in the early stages, including the US National Science Foundation, the European Commission, and the National Aeronautics and Space Administration (NASA), as indicated by acknowledgment in the scientific literature. Nonetheless, depending on the funding rules, this does not prevent scientists directly involved in the research from applying for patents under their own names. Prominently, while governments provided most funding in the 2010s, philanthropies have contributed more since 2020 (as shown in Fig. 1).Footnote 34 These include funding for Innovative Climate and Energy Research from Bill Gates and research on sea spray geoengineering and marine cloud albedo from the Maj and Tor Nessling Foundation.Footnote 35 In addition, private start-ups, in particular those from the US and Israel, are pursuing a proprietary approach to commercialising the patents.Footnote 36
Funding source of SRM activities (2007–2024).
However, accurately mapping the patent landscape in SRM is challenging due to the lack of a specific patent class aligned with the existing classification of different forms of geoengineering technology. This may lead to ambiguity and disinformation.Footnote 37 Existing literature has used either a keyword-based methodologyFootnote 38 or a combination of major patent subclasses and keywordsFootnote 39 as a patent search strategy. These data are not explicitly related to the research focus on risk regulation. For instance, the latest patent landscape analysis has identified the most patented technologies in relation to SRM, including battery energy storage, biofuel production, CO2 capture, CO2 capture and/or removal, control and management of climate disasters, energy recovery, oil processing with organic materials, production of chemicals using catalysts, production of chemicals using catalysts and recycling of catalysts, solar energy, photovoltaic, preparation of compounds containing monosaccharide radicals, wind turbines.Footnote 40 This list encompasses a wide range of technologies, including dual-use technologies that serve SRM and other purposes – for instance, a patent (GB202305479D0) for a high-altitude aerostat with a large surface area can be used both for SRM and for airborne solar power generation.Footnote 41 The scope of technologies is also very broad, including other climate mitigation fields that are less controversial (e.g., carbon removal and renewable energy). If this is taken as a basis for discussing the possible prohibition of patenting for SRM technologies, the broad scope would likely attract unnecessary objections.
Therefore, this article employs a refined patent searching strategy, as outlined by Ramos and Santos (2025).Footnote 42 Among all sub-technology areas of SRM, only SAI was the focus of the patent search. The author first identifies the most relevant patent classes (A01G15/00 or B64D47/00) based on International Patent Classification (IPC) and then uses the keywords “stratosphere” and “aerosol” to identify patents relevant to SAI technology within the classes. Then “solar radiation management” was added as an additional keyword. We then retrieved cited and citing documents. This first yielded a total of 1,425 patents. A stringent data-cleaning strategy was adopted to retain patents that appear directly related to SRM. After reviewing keywords, 397 patents remain.Footnote 43 Initial data show that private actors have actively pursued SRM patents for over two decades. After patent family consolidation, the US is the country of origin for half of SRM patent applications (148), followed by China (50) and Germany (21) (Fig. 2).
Country of origin of the patent applicants for SRM patents.
In terms of applications, Fig. 3 shows a drop from 2014 to 2016, but since then the number has gradually increased. It is worth noting that the patent application numbers appear to reflect the trend in R&D funding for these technologies in certain respects. Still, more rigorous empirical studies are needed to confirm the correlation and the lead time, if any.
Annual SRM patent application (2006–2025)Footnote 44 .
Fig. 4 shows the distribution of IPC classes for SRM patent applications. It is worth noting that this patent search approach has significantly affected the distribution. Only the two most relevant IPC classes are selected as the basis for the patent search. The first is A01G (horticulture; cultivation of plants, forestry, and irrigation), which includes A01G15/00 devices or methods for influencing weather conditions. This is because many weather modification techniques were first developed for use in horticulture and agricultural contexts, such as cloud seeding to enhance or prevent precipitation. The second is B64D (Equipment, Systems, or Components for Aircraft), in particular B64D47/00 (equipment not otherwise provided for). These are mainly auxiliary equipment or systems mounted on aircraft for specialised purposes, such as meteorological instruments. It is not surprising that the two classes are the most concentrated. However, the forward and backward patent citations have also led to a more comprehensive patent classification, in particular patent applications broadly classified under B64 (Aircraft; Aviation; Cosmonautics), such as various types of balloons under B64B, aeroplanes and helicopters under B64C, and cosmonautics vehicles and equipment under B64G.
IPC distribution for SRM patent application (2006–2025)Footnote 45 .
However, it should be noted that the above data can be considered only a proxy for the patenting landscape due to the fuzzy boundaries caused by the lack of a clear concordance between patent classification and SRM technology classification. A more rigorous patent landscape analysis is needed for any future regulatory arrangements regarding SRM technologies.
2. Grounds and proposals for excluding patents on SRM technologies
The extant literature has mapped regulatory pathways to restrict patents for SRM technologies. These proposals include ex ante measures and ex post measures. The former concerns the grounds for granting a patent and excludes certain objects from patentability.Footnote 46 Ex post measures allow patenting in the first place, but they either restrict the exercise of patent rights (such as compulsory licensing, Crown use, and a Doha-like DeclarationFootnote 47 ) or encourage innovative and more efficient uses of patent rights, such as patent pools and a march-in right.Footnote 48 Most discussions on the intersection of patents and other areas of law requiring certain restrictions on patents have focused on ex post measures. For instance, compulsory licensing (licensing without the patent holder’s consent, as outlined in TRIPS Article 31) is available for pharmaceutical patents to promote access to medicines, but its applicability to environmentally sound or clean technologies is unclear.Footnote 49 At the 2008 International Conference on Carbon Abatement Technology Transfers in Beijing, India and China proposed extending compulsory licensing under the TRIPS Agreement to environmentally sound technologies, which previously covered only pharmaceutical products.Footnote 50 Similarly, discussions of voluntary efforts to pool technologies often assess the effectiveness of previous cases of ecological innovation. Unfortunately, evidence shows that such pools have not increased the diffusion of patented technologies to an adequate level to address the climate crisis.Footnote 51 In general, ex post measures are less controversialFootnote 52 and often only require clarification of the TRIPS Agreement or domestic implementation.
To advance relevant discussion on SRM patenting in response to the call by the African Groups, Mexico, Colombia, and some civil society organisations for a non-use mechanism for SRM,Footnote 53 this article focuses on the grounds and proposals for prohibiting patents for SRM technologies. This is because proposals to adjust the exercise of patent rights within the TRIPS Agreement are inadequate not only to remove patents as a barrier to equitable access to certain technologies but also to control potentially harmful technologies, such as SRM. Even though relevant ex ante proposals may not seem feasible under the current TRIPS Agreement, recognising the barriers to these proposals is the starting point for exploring opportunities for structural change.
a. Ordre public and explicit morality exclusions
Some limits to patentability have been put in place to exclude inventions contrary to the Ordre Public and morality. In the European Patent Convention (EPC) Article 53(a), it is provided that: “European patents shall not be granted in respect of inventions the commercial exploitation of which would be contrary to ordre public or morality.” While one interpretation is that patent protection is denied to immoral inventions, no matter how novel or inventive the invention may be, it is difficult to enforce this rule. So far, morality in this article has been narrowly interpreted concerning Catholic and Christian beliefs, the moral status of embryos, and the morality of patenting animals and parts of the human body. Even though it has been repeatedly pointed out that extractive IP might impede the diffusion of climate-related technologies and reinforce the North-South divide, the moral case for excluding patents in this regard has not been widely accepted.Footnote 54
Admittedly, morality, in general, is difficult to substantiate in individual cases. It has been pointed out that when a patent examiner or a court assesses whether an invention is so immoral that it should not be granted patent protection, they address the complex interaction between IP law and philosophy.Footnote 55 Nonetheless, the challenging nature of this exercise does not mean it should not be undertaken, particularly when there is higher uncertainty about the consequences of deploying technologies such as SRM. Explicit morality exclusions have been proposed as a “policy lever” to tailor patent law to its overarching utilitarian objective of promoting socially beneficial inventions.Footnote 56 On the basis of the risks of SRM technologies or the risks arising from SRM governance, the following moral arguments can be made for the categorical exclusion of the patentability of certain SRM technologies.
SRM technologies are not just any emerging technologies but “ambiguous” technologies for which potential risks and risk management capacities are uncertain, unknown, or even unknowable. In this regard, the principle of precaution emphasises the importance of avoiding actions that may cause unacceptable harm, even in the absence of scientific proof of causation. In this respect, SRM entails irreducible uncertainty in assessing societal consequences (i.e., uncertainty that cannot be resolved through additional empirical research or ethical analysis).Footnote 57
Potential SRM risks have been discussed in the literature. These include the moral hazard problem that SRM, as a cheaper alternative, will deter continuous climate mitigation commitments and, therefore, potentially be detrimental to preventing global warming, and further enable fossil-fuel-based business as usual.Footnote 58 In addition, because SRM provides relatively inexpensive, rapid temperature control, it may lead to the free-driver problem, and the low cost of deployment may lead to a risk of controversial unilateral intervention.Footnote 59 This possible, inevitable overuse of unilateral deployment may further exacerbate equity concerns at the global level – that decisions over SRM experimentation and deployment would not be made in a globally inclusive, equitable and transparent manner, thereby deepening power disparities and contravening the value of non-domination.Footnote 60
However, it is unlikely that these exclusions will be agreed upon through amending the TRIPS Agreement due to its contravention of the principle of technology neutrality (see Section II.3). Therefore, the primary challenge for implementing explicit morality exclusions is whether and how to include such a policy lever in national patent laws, bearing the risks of being sued for TRIPS violations.
b. Harmful inventions and immoral patents
In addition to explicit morality exclusions, Ned Snow argues that certain intellectual creations should not be patented simply on the grounds that they are harmful to society; such moral limitations can be justified by traditional IP theories.Footnote 61 This can be a powerful argument to exclude patents for certain SRM technologies, but there are a couple of predicaments in applying this moral argument to SRM. There is a substantial difference between harm (actual damage or adverse effect) and risk (likelihood and severity of a potential adverse outcome). Harm from SRM can only come from deployment. Therefore, this morality can only be applied once the harm is proven after deployment. However, due to deep concern about the unknown consequences of its deployment, many argue that SRM should not be deployed. If patents are granted before harm is proven, once deployment is permitted, commercialisation funded by private investment will flourish. These patents will become the cornerstone for profiting from SRM, like patents for other technologies. Then, it is difficult for this moral argument to prevail in the face of a positivist argument based on the TRIPS principle of technology neutrality. Instead, the patent rights holders will argue that liability is a separate issue and should be regulated outside of the patent law. Alternatively, if patents are granted to SRM technologies but deployment has not yet been permitted, patents can still be used as leverage to promote deployment. This is because Article 27.1 of the TRIPS Agreement also provides that “patent rights [shall be] enjoyable without discrimination as to … the field of technology,” and deployment can be considered a necessary way to enjoy the patent rights.
Notably, harmful inventions are not a regulatory vacuum. Existing international prohibitive and restrictive regimes ban or restrict the deployment of technologies, such as chemical weapons, biological weapons, weather modification technologies, anti-personnel landmines and substances that deplete the ozone layer, based on human rights, environmental concerns or customary international law.Footnote 62 As discussed above, the existing SRM regulations only cover the deployment stage; little has been discussed about monopolistic control of relevant technologies at the early stage of R&D via patents.
Focusing solely on non-deployment without considering patents may not be sufficient for SRM, and a non-use mechanism should account for both patent prohibition and deployment. There are inherent dual effects of SRM, which could reduce climate risks while introducing new unknown risks. Current regulatory principles do not have clear guidance on how to address dual effects.Footnote 63 Due to their uncertainty, governance systems are indeterminate regarding ambiguous technologies such as SRM, which means existing norms and rules do not imply default solutions,Footnote 64 despite a relatively strong case for the non-use mechanism of SRM.Footnote 65 Due to such indeterminacy, there could be two scenarios in which patents can hinder the decision against deployment if the regulation focuses only on deployment. For one thing, if there is a decision on non-deployment (more likely at the national level), there is likely an argument that the ban on deployment hinders the right to equally enjoy patent rights without discrimination, in line with the TRIPS principle of technology neutrality. For another, if the deployment of SRM is on the agenda, monopolistic patents will make it less affordable to the least developed countries that may need it most, just as patents have hindered equitable access to pandemic products.Footnote 66
c. Provision of SRM as a global public good?
The most influential principles proposed for SRM governance are the Oxford Principles, which set out five principles for responsible conduct of geoengineering research. The first of these principles is “Geoengineering to be regulated as a public good.”Footnote 67 Public goods have two features: non-excludable and non-rival. Once provided, no one can be excluded from consuming them, and one person’s consumption of them does not prevent others’ consumption.Footnote 68 Scientific knowledge is inherently a public good. If not restricted, no one is excluded from benefiting from knowledge and information; one person’s use of knowledge does not interfere with others’ use. However, the patent system is the key instrument for creating exclusivity for scientific knowledge in regulatory capitalism. Commodified science has exchange value in the market, and its enjoyment is conditional on payment. One of the tragedies of commodification under capitalism is the undersupply of global public goods, such as those related to climate change and biodiversity.Footnote 69
If the harms of SRM inventions can be confirmed, an international prohibitive and restrictive regime can be justified, following the international law and global governance of other harmful inventions and materials.Footnote 70 However, the above unknown risks and irreducible uncertainty associated with the ambiguous technology also pose challenges for assessing whether they are “public good” or “public bad” in the first place. This ambiguity is further complicated by the fact that some SRM technologies can also be used for climate-related purposes. For instance, the patent US20090032214A1 “System and Method of Control of the Terrestrial Climate and its Protection against Warming and Climatic Catastrophes Caused by Warming such as Hurricanes”Footnote 71 (abandoned) designs a system for controlling the Earth’s climate by using aircraft to release sun-shading aerosols into the upper atmosphere to create a controlled cooling effect similar to a mini nuclear winter. In addition to solar radiation management, the same technology, using sun-shading aircraft fuels and dispersing sun-shading sprays in the areas of and around tropical storms (or potential storms) and along their paths, could also be used to reduce the intensity and destructive force of tropical storms, hurricanes, typhoons, cyclones, and other weather catastrophes. Therefore, the same invention can be used both for SRM with unknown risk and for preventing climate catastrophes caused by extreme weather such as hurricanes. For this latter purpose, the invention may still be considered a public good with potential benefits and should be patentable. However, the patent system needs to provide greater flexibility to facilitate knowledge diffusion. Another type of dual-use is of inventions that can be used for both SRM and ordinary purposes. For instance, it is suggested that existing aircraft can be used for stratospheric aerosol injection.Footnote 72 At the minimum, inventions related to the ordinary use of the aircraft should be eligible for a patent.
This dual-use challenge arises because the patent system focuses on the technical merits of an invention (i.e., one that is novel, non-obvious and solves a technical problem), and the intended purpose(s) of use are not issues patent examiners may consider when assessing patentability. Should the proposal to prohibit SRM patents proceed, the dual-use challenge of identifying the specific SRM use and associated risks must be addressed. A proper verification system must be established for the dual-use claims, as it has long been pointed out that “to enhance patent eligibility, geoengineering technology inventors could try to conceal, suppress or misrepresent adverse information about the risks and impacts of such technologies while at the same time portraying such technologies as the solutions to climate change.”Footnote 73
As pointed out by Drahos, the provision and distribution of public goods are deeply affected by the degree of excludability of those goods and the regulatory context of that excludability.Footnote 74 The patent system, as the primary regulatory instrument for defining the scope of knowledge exclusivity, remains the primary regulatory mechanism for the SRM regulation.Footnote 75 The possibility of certain inventions can be both SRM and climate dual-use technologies, thereby distinguishing SRM technologies from purely immoral inventions that should simply be prohibited. This requires a more balanced and precautionary approach to a patent system that includes ex ante dual-use risk assessment (Section III).
d. The proposal of an international non-use agreement on SRM
With the intensification of the debate over potential deployment of SRM as a last resort when facing inadequate climate mitigation commitments, a group of scientists proposed an SRM non-use agreement in early 2022.Footnote 76 It calls for a pre-emptive prohibition of SRM at the international level, with five proposals: no public funding for SRM research, no outdoor SRM experiments, no patents on SRM technologies, no SRM deployment, and strengthening multilateral governance. Patents were specifically emphasised: “Intellectual property rights for SRM technologies would not be granted, reducing the potential for commercial incentives to drive their development.”
The proposal reflects a precautionary approach – indeed, the grounds discussed above can all constitute the basis for taking such a precautionary approach to prohibit patents on certain SRM technologies instead of the ex post restrictions on patents. Including patents in these four major prohibitions also indicates the importance of the role that patents play in SRM technologies. As discussed above, it defines the “ownership” of respective SRM technologies and remains the major institution that distributes benefits from SRM if deployed.
However, it has been pointed out that, although commentators have opposed deployment for SRM, none of them is underpinned by a process through which a prohibition could feasibly and effectively be implemented and enforced.Footnote 77 This is also true for the proposal to eliminate patents for SRM – because of its potential conflict with the TRIPS principle of technology neutrality, national patent laws may not adopt it.
3. TRIPS principle of technology neutrality
Despite all the proposals to prohibit patents for SRM technologies, particularly the non-use agreement, it seems that the chances are slim that any of them will be implemented as a patent regulation specific to SRM technologies. Article 27.1 of the TRIPS Agreement provides that “…patents shall be available for any inventions, whether products or processes, in all fields of technology, provided that they are new, involve an inventive step, and are capable of industrial application.” This is often referred to as the principle of technology neutrality, under which national patent laws may not differentiate in their treatment of patentable subject matter on the basis of the technology field. According to this principle, SRM technologies should not be excluded from patenting as long as they fulfil the patentability requirements.
For the proponents of the International Non-use Agreement, one challenge is the literal, formalistic approach to interpreting this principle. Existing WTO jurisprudence seems to support this approach. In India – Patent Protection for Pharmaceutical and Agricultural Chemical Products, the US was concerned about the absence in India of either patent protection for pharmaceutical and agricultural chemical products under TRIPS Article 27, and India’s argument based on “legitimate exception” was not supported during the appeal.Footnote 78
In the Canada – Patent Protection of Pharmaceuticals (DS 114), a case concerning a regulatory exception applicable to certain types of inventions, the stockpiling exception was found to violate the technology neutrality principle.Footnote 79 This exception in question allowed generic manufacturers to produce and stockpile patented drugs six months before patent expiry, which was considered an unreasonable prejudice to patent holders’ enjoyment of their exclusive rights. In the issue of “discrimination as to field of technology,” the panel acknowledged that “discrimination” can be broadly defined as “results of the unjustified imposition of differentially disadvantageous treatment.”Footnote 80 However, with respect to TRIPS Article 27.1, it decided to defer defining the term at the outset and to focus on the issues raised; in other words, defining it “to the extent necessary to resolve those issues.”Footnote 81 In this case, while justified differential treatment was impliedly allowed, the scope of this justification found was very narrow. It confirmed that the implementation of the TRIPS Agreement is “in principle neutral vis-à-vis societal values.”Footnote 82 It interprets the important principle of Article 8.1 of the TRIPS AgreementFootnote 83 that “none of the public policy considerations, referred to in the first half sentence, could be invoked to justify measures which were inconsistent with provisions of the TRIPS Agreement.”Footnote 84 Therefore, even though a measure may be permissible as necessary to achieve certain well-defined societal values, if specific provisions in TRIPS do not refer to those values, the exception cannot be justified under TRIPS. Following this approach, exclusion of subject matter from patentability based on a technology category may be considered a violation of the TRIPS principle of technology neutrality.
Beyond the WTO jurisprudence, there are theoretical arguments against the principle of technology neutrality. First and foremost, treating all technologies equally is legal fiction, as, in reality, technologies are never equal. Different categories of technologies have different life cycles (IT technologies are much shorter-lived than pharmaceuticals), different levels of intellectual effort and skills required, different financial costs and different prospects for commercialisation. Regarding SRM-related technologies, they also have different climate implications and different risks when implemented. The principle of technology neutrality in TRIPS benchmarks technologies solely on the basis of patentability (novelty, inventive step and industrial applicability) and ignores other differences, thereby failing to be neutral and favouring certain values or policy objectives over others. Second, the TRIPS Agreement entered into force in 1995, when certain new technologies, such as AI and CRISPR gene editing, were not on the horizon, and challenges like climate change were not seriously addressed in IP law. While technology neutrality is often said to sustain the longevity of laws, there is a risk of automatically extending old rules to new technologies that were unknown when the law was enacted.
Exceptions to this principle are provided in TRIPS Articles 27.2 and 27.3 (see also Section II.2.a). Article 27.2 is particularly relevant for possible SRM inventions to be excluded from patentability as it allows WTO members to exclude inventions from patentability within their territories when it is necessary to protect the ordre public or morality, including to protect human, animal, plant life or health, or to avoid serious prejudice to the environment. The key question is whether excluding SRM inventions categorically from patenting would be allowed by TRIPS Article 27.2 as “necessary to avoid serious prejudice to the environment.”
A doctrinal approach would focus on such an exception can be justified as “necessary” to avoid “serious environmental prejudice,” but the TRIPS Agreement provide little guidance on how each Member State should determine whether excluding particular inventions from patentability qualifies as “necessary.”Footnote 85 Weissman argued that existing WTO jurisprudence interpreting GATT Article XX would provide persuasive authority for the requirement of necessity under Article 27.1, implying that patent restrictions would be considered “necessary” only if no alternative measure less inconsistent with the TRIPS Agreement is available to achieve its goals.Footnote 86 GATT Article XX dispute outcomes suggest that if this approach is taken, there is little chance of winning the necessity test.
By contrast, the regulatory discretion approach recognises the regulatory sovereignty enjoyed by states under the TRIPS Agreement when WTO members implement their patent systems. In particular, the Declaration on Patent Protection acknowledges that:
Article 27 of the TRIPS Agreement does not prevent states from reasonably differentiating between fields of technology according to
– the characteristics inherent in the technology at issue and
– the state’s public policies pertaining to the sector at issue.Footnote 87
This approach confirms sovereign states’ discretion to shape a patent system that best suits their technological capabilities and social, cultural and economic needs and priorities.
These two approaches reflected two different conceptions of state autonomy under international IP law. The doctrinal approach focuses on whether the exception is permitted within the scope of the exceptions, which presumes that deviations from TRIPS obligations must be explicitly justified by exceptions and shifts the burden onto the state to demonstrate that a measure is “necessary,” a stringent requirement to fulfil. By contrast, the regulatory approach is grounded in the logic of state sovereignty. Unless the TRIPS Agreement expressly prohibits a measure, WTO Members can retain their regulatory sovereignty to design their domestic patent systems as they see fit. Since the TRIPS Agreement was aimed at setting minimum standards instead of a detailed and exhaustive treaty, regulatory space retained by the WTO members should be treated as a default starting point. Therefore, exclusion of SRM technologies from patentability does not need to be granted through exceptions. Nonetheless, it is admitted that the possibility of being sued at the WTO may have a chilling effect on countries considering excluding SRM technologies from patenting. One way to address the challenge is to explore more structural change from outside the TRIPS Agreement.
III. Making risk regulation a patent-related issue
The patent system plays a key role in the commodification of knowledge through the granting of proprietary rights, and the principle of technology neutrality means that the patent system serves as a default regime for regulating SRM technologies. It has been challenging to incorporate alternative norms into current IP law. This section first considers incorporating risk regulation into the patent system and then proposes a sui generis SRM regulatory framework that includes a prohibition on certain SRM inventions from patenting. This adds to the regulatory toolbox, as the current proposals are either within the existing IP legal framework and do not consider the potential harm of SRM,Footnote 88 or a blanket prohibition that does not properly address the SRM-climate dual-use challenge.Footnote 89
1. Ex ante risk regulation as a patent-related issue
This section proposes a theoretical framework that can incorporate ex ante risk regulation into the consideration of excluding SRM technologies from patenting. In their classical law-and-economics analysis, Guido Calabresi and Douglas Melamed distinguished between property rules and liability rules as regulatory tools to protect entitlements: property rules discourage nonconsensual takings, while liability rules permit nonconsensual takings in return for payment of damages. Although patent law certainly includes secondary liability rules for payment of damages in cases of infringement, the patent system is primarily a system of property rules. Although liability rules compensate for potential harm from deployment and are part of an independent regulatory regime,Footnote 90 in the current system design, there is very little room for liability rules to be part of the regulation of the entitlements to the technologies in the first place. However, there are arguments from law and economics that can support the incorporation of risk regulation in the patent system.
Following the law and economics theory, one can revisit the suitability of property rules and liability rules as proposed by Calabresi and Melamed. They argue that the criterion for choosing between property rules and liability rules should be based on transaction costs: property rules should apply when transaction costs are low and the parties can bargain with one another, while liability rules are best applied when transaction costs are high, and bargaining is impossible or difficult.Footnote 91 If SRM are deployed, the harm to the Earth will be unpredictable and irreversible. Therefore, it is not a bargaining between a group of scientists and the incumbent TRIPS Agreement, or between more vulnerable communities and technology tycoons or superpowers (admitting that bargaining can be difficult enough), but one between human beings and the only habitable planet.
This also leads us to recognise the limits of regulatory instruments primarily devised for regulatory capitalism, including the patent system, which prioritises economic logic and is less responsive to non-economic, ecological values and issues.Footnote 92 The assumption underlying market-based regulation is that if a market price cannot reflect the value of something – that is, if it is not a commodity with exchange value – the value is presumed to be zero. In this case of SRM regulation, even though the Earth cannot respond to the market, the potential risks are existential and should not be ignored. Indeed, if non-market values are considered, the regulation of SRM constitutes a situation where “bargaining is difficult,” and liability rules should not only be used as secondary rules. Further, other laws and regulations can substantiate ecological values through prohibitions on operation, cancellation of qualifications, fines or damages. This means an ex ante risk assessment (including dual-use risk assessment) in patent law could be beneficial for rights holders because there is enormous uncertainty about the harm from SRM deployment – should ex post liability be the only way to compensate for the harm, rights holders may not be able to afford the cost.
Once risk regulation is incorporated as ex ante risk considerations into the decision on patent granting, it opens the possibility for further discussion of whether they should be elements in determining whether patents should be granted for certain SRM technologies, or whether they should be part of the patentability assessment for individual applications. There will also be scope for discussions about which risks should be incorporated into such decision-making.
2. A Sui generis regulatory system for SRM
While there are extensive demands for the potential future governance of SRM, the current governance landscape is fragmented. As pointed out, the most challenging issue in SRM implementation is not technical feasibility but institutional feasibility, which can enable an implementation strategy that optimises benefits and minimises risks.Footnote 93
Various international regimes are segregated by subject matter, including climate change, the ozone layer, the sea and the Antarctic, and SRM decision-making will also be subject-specific, focusing on when, where and how SRM technologies will be used. In international law, there are numerous specific elements from which governments could draw implications to constrain or prohibit SRM.Footnote 94 This also leads to an alternative, technology-specific approach in patent law as an exception to the principle of technology neutrality (Section II.3). Such an approach would allow tailored regulation of certain technologies. Some exceptions in TRIPS Article 27.2, as exceptions to the principle of technology neutrality, can be considered a technology-specific approach. The challenge is that, after the TRIPS Agreement was concluded with its original exceptions, it is almost impossible to adopt new exceptions, as this would require renegotiating the treaty.
Following this approach, a sui generis regulation for SRM technologies is proposed. After comparing the innovation structure of SRM, biotechnology and nuclear technology, a report concluded that SRM should be treated like atomic energy with a sui generis patent system.Footnote 95 While risk regulation was not the only consideration for such a proposal, as there are also issues of national security implications, public controversy and the need for global coordination in innovation, a sui generis system outside of the TRIPS Agreement provides an opportunity for the abovementioned risk considerations to be thoroughly examined in the institutional design.
As the first proposal for a sui generis regulatory system for SRM technologies dates back to 2010,Footnote 96 a sui generis approach is not new. However, recent proposals for a non-use agreement and the cautious attitudes of developing countries and the EU have leaned towards a sui generis patent prohibition system. Implementing a technology-specific approach to design a dedicated patent system for SRM technologies would still face the challenge posed by the TRIPS Agreement’s principle of technology neutrality in the normative ordering, as discussed in Section II. However, discussions and lessons in the last decade do offer some new insights. Conceptually, a sui generis system only means it is a special system for SRM technologies, but the emphasis should be put on how special treatment can be designed within the system. Considering the ambiguity of SRM technologies and the intersectionality of the regulation of innovation and dissemination of SRM technologies, and regulation of risks, harms, funding rules, deployment and relevant security and equity concerns, what is required is not necessarily a sui generis patent system for SRM technologies, but a sui generis regulatory system for SRM that has special consideration for innovation stimulation and diffusion (including the possibility of prohibiting patent grants for certain SRM technologies).
A unitary sui generis system for SRM may also avoid the drawbacks of the International Non-Use Agreement on Solar Geoengineering. As discussed in Section III.1, it needs to justify why certain SRM technologies should be treated as exceptions to the TRIPS principle of technology neutrality, not only for traditional moral considerations but also for practical considerations acceptable to stakeholders, including patent applicants and rightsholders. The law and economics analysis suggests that incorporating ex ante risk regulation as an alternative to property rules (i.e., patents) will benefit investors and patent applicants, incentivising them to undertake due diligence and preventing them from ex post liabilities that may be imposed by other laws and regulations at both international and national levels. If the exclusion of patentability is not accepted as a legitimate exception to technology neutrality, there are legal consequences based on the rules of liability. When the interests of rightsholders are aligned in this way to consider property rules and liability rules as alternatives to SRM technologies, a sui generis regulatory system may face fewer hurdles under the TRIPS Agreement’s principle of technology neutrality.
The next question is how to integrate special considerations. Suppose the aim is a binary system that categorises relevant patents and prospective applications as either patent-permissible or patent-prohibitive based on risk assessment. The key question is where to draw the threshold and on what basis. However, the categorisation issue may take a long time to negotiate. Alternatively, a layered system could be introduced based on the risk assessment, ranging from full exclusive rights to no exclusive rights. This is also where the ex post restrictions can be integrated.
Furthermore, the sui generis regulatory system should focus more on potential strategic implementation than on the expected outcome of excluding patents from SRM technologies, as proposed in the International Non-Use Agreement on Solar Geoengineering. For instance, it can refer to experiences of international prohibitive and restrictive regimes on certain risky or harmful activitiesFootnote 97 to create synergies between the patent system and other laws and regulations. In addition, the dual-use issue needs to be taken into consideration – if there is a patentable invention that can be used for SRM and other purposes, such as aerostat, the sui generis regulatory system needs to further coordinate with the ordinary patent system regarding exclusivity for dual-use technologies.
IV. Conclusions
The article examines proposals and challenges associated with prohibiting patents for certain SRM technologies, primarily based on risk considerations. The patent system plays a key role in the commodification of SRM technologies. It remains the core system for relevant stakeholders to profit commercially from SRM technologies once deployment is permitted. Also, patents may serve as a basis for advocacy for deployment, as they can be considered a necessary way to enjoy the rights conferred. Despite various grounds for prohibiting the granting of patents for SRM technologies, the implementation of such exclusion by national patent laws may nonetheless contravene the TRIPS Agreement’s principle of technology neutrality.
Recognising the dominant position of IP in normative ordering, this article proposes considering ex ante risk regulation as a patent-related issue, grounded in an economic analysis of law, and best implemented through a unitary sui generis regulatory system for SRM. Calabresi and Melamed argue that property rules should apply when transaction costs are low and the parties can bargain with one another; liability rules are best used when transaction costs are high, and bargaining is impossible or difficult. If SRM is deployed, the harm to the Earth will be unpredictable and irreversible, so transaction costs will be very high if non-economic and ecological values are not considered. Instead of relying on exclusions from patentability mainly based on moral considerations, the proposed ex ante risk consideration in patent law is based on a refined economic analysis of law – should the liability legitimately imposed by other laws on the deployment of SRM technologies become so high, the law and economics analysis itself could lean toward incorporating ex ante risk considerations into patent law. The proposed sui generis regulatory system for SRM differs from a sui generis SRM patent system. It considers the intersectionality between the regulation of innovation and the dissemination of SRM technologies, as well as risk regulation. As a unitary system dedicated to SRM, it may be open to broader regulatory issues, including those related to patents, funding rules, deployment, and relevant security and equity concerns.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/err.2025.10066
Acknowledgment
The author thanks Professor Peter Drahos and Dr Hazel Moir for their comments and suggestions on the early draft of this article.
Open access
