First Comes the Science: What Gene Editing's Legitimacy Gap Reveals for SRM

This blog post is one in a series drawing out lessons for solar radiation modification from efforts to govern other emerging technologies. The goal is to seek insight into how institutions and societies can best prepare for the complex task of managing powerful and disruptive technological innovations. The introduction to the blog series is here.

Proponents of gene editing and SRM often frame both as pragmatic, albeit extreme, solutions to urgent global challenges. Gene editing is positioned as a means to treat human and non-human genetic diseases, enhance crop resilience, and potentially eradicate vector-borne illnesses such as malaria. SRM, particularly stratospheric aerosol injection (SAI), is discussed as a potential method to reduce extreme climate impacts while emissions reductions continue. In both cases, the primary appeal lies in the potential to address complex global problems through targeted scientific intervention.

At the same time, the story of gene editing exposes a clear governance dilemma: scientific advancements frequently outpace the development of oversight institutions. Public deliberation, regulation, and coordination tend to evolve slowly, whereas technological progress can occur rapidly. As a result, governance debates often arise only after technological capabilities have already expanded.

SRM is at an early enough stage of consideration that there is still scope for anticipatory governance - that is, the building of governance capacity before the rush of innovation and events forces reaction. Our intent across the series of which this blog post is a part is to adopt a readiness lens. Through such a lens, it becomes clear that institutions are frequently pushed into a reactive posture, responding to controversies after they arise rather than shaping technological trajectories in advance. Governance follows disruption rather than guiding it. When that happens, a central challenge becomes legitimacy. Public trust becomes fragile when technologies appear to move forward without clear oversight, meaningful transparency and participation, or shared agreement about their purposes.

Recent developments demonstrate how rapidly this dynamic can manifest. As SRM transitions from theoretical discourse to policy consideration, the trajectory of gene editing provides a valuable comparative case. This example illustrates how swiftly disruptive technologies can reshape governance debates and how challenging it is to restore legitimacy once the perception arises that decisions outpace collective societal deliberation. Recognizing these parallels highlights a broader issue that transcends individual technologies. As the capacity to intervene in planetary and biological systems expands, the central challenge becomes not only the efficacy of these tools, but also whether governing institutions can evolve rapidly enough to maintain public trust.

From laboratory to governance challenge: The rise of gene editing

Genetic modification has existed in various forms for decades. Gene editing involves targeted modification of genetic material, allowed by development of a new genetic modification toolkit. We are focused here particularly on CRISPR.  CRISPR, which stands for Clustered Regularly Interspaced Short Palindromic Repeats, has made gene editing faster, cheaper, and more accessible than earlier methods. The technology and approach allows scientists to delete, insert, or modify genetic material with a precision and ease that previous tools could not match. In only a few years, CRISPR has dramatically expanded what scientists can do with DNA, forcing governments and international institutions in turn to grapple with ethical and regulatory questions that had barely existed a decade earlier. Oversight systems built for slower, more centralized science suddenly faced a world where many more actors could move quickly, in more contexts, with less friction. 

Like SRM, gene editing encompasses multiple forms of intervention, each with different scales, risks, and implications for governance. Somatic editing changes non-reproductive cells in an existing person, so the edits are not inherited. This category aligns, at least conceptually, with many forms of medical treatment and has often been treated as less ethically fraught. Germline editing, by contrast, involves embryos, eggs, sperm, or early developmental cells, meaning changes could be inherited by future generations. This raises questions that standard medical oversight isn’t yet designed to settle: consent across generations, impacts on human inheritance, and the risk that social inequality becomes biologically reinforced. Public acceptance tends to fracture along those lines. The same tool can be seen as medicine in one use case, and as social engineering in another, which makes governance harder because the boundaries are not technical; they’re political and moral.

CRISPR also enabled gene drives, a different category of intervention. Where standard inheritance passes traits to some offspring, a gene drive spreads them through entire populations across generations. In practice, CRISPR-based gene drives can be designed to address malaria or invasive species management by spreading traits through mosquitoes, rodents, or other organisms, potentially suppressing populations, reducing disease transmission, or protecting ecosystems.

The parallel is sharpest with SAI. Like gene drives, SAI would produce effects that cross borders, affect populations that had no say in the decision to act, and raise fears about what happens if the intervention is stopped once it has begun. Gene drives can have ecological effects that move across borders, create contested questions of consent, and raise fears about irreversibility or loss of control once released into dynamic systems.

The governance problem: Capability moves first, institutions react

Across these categories, a patchwork of norms emerges that may occasionally deter risky behavior but lacks consistent cross-border enforcement. Jurisdictional differences in regulation enable actors willing to test boundaries to exploit permissive gaps. While most researchers act responsibly, voluntary and inconsistently applied governance can inadvertently reward those who do not. When outliers engage in questionable practices, the reputation of the entire field is affected.

The rapid adoption of CRISPR exemplifies a recurring pattern: when technological capabilities expand more quickly than oversight mechanisms, governance becomes reactive. CRISPR was first demonstrated as a practical tool in 2012 and was rapidly adopted by laboratories globally. Regulatory systems depend on committees, legislation, and deliberation, whereas scientific progress advances through protocols and replication. These timelines are often misaligned.

One result has been governance fragmentation. There is no unified global regulator for human genome editing. National rules vary: some countries allow gene-editing research on embryos under specific conditions while still restricting heritable clinical use, while others set tighter limits. Professional bodies issue guidelines that shape what journals publish and what funders will support, but they cannot enforce compliance beyond those channels. At the international level, oversight relies on a mix of bioethics and human rights instruments, professional guidelines, and guidance from bodies like the World Health Organization and academies of science. The Council of Europe's Convention on Human Rights and Biomedicine, known as the Oviedo Convention, draws a bright line against heritable genome editing among its signatories. International bodies can coordinate norms, but they tend to act slowly and often lack binding authority. This creates expectations and guardrails, but it also leaves gaps.

These governance gaps became most apparent in 2018, when He Jiankui announced the use of CRISPR to edit human embryos, resulting in the birth of twin girls. This event prompted a global reckoning and served as a governance stress test that the system failed—not due to a lack of awareness of ethical risks, but because existing mechanisms were insufficiently robust, consistent, or legitimate to prevent boundary crossing. In response, the World Health Organization convened advisory groups and emphasized that clinical germline editing should not proceed without stronger governance. Many scientists condemned the work, and several countries reviewed or tightened their regulations. The sequence of events is significant: institutional responses were reactive to the incident, rather than the result of proactive governance readiness.

This is the governance pattern that SRM risks repeating. In both domains, the moment of crisis is not the first time governance becomes relevant. It is simply the first time the wider public is forced to pay attention. By then, legitimacy is already fragile.

Legitimacy: When trust becomes the constraint

Gene editing shows what happens when a technology outpaces the institutions meant to oversee it. Rather than governing a trajectory, those institutions end up responding to crises. The pattern is clear in the gene-editing timeline: rapid expansion after 2012, governance responses that came late, and public trust that eroded in the gaps. Potential or even likely technical success is not enough to justify a technology that touches shared moral ground. Even when a procedure works, people ask who authorized it, whose values shaped the decision, and who bears the risk if something goes wrong.

The 2015 International Summit on Human Gene Editing distinguished between technological capability and societal consent. The summit concluded that advancing toward clinical germline editing necessitates not only demonstrated safety and efficacy, but also broad societal agreement regarding the appropriateness of such applications. Legitimacy was identified as a prerequisite rather than a matter of communication. Once the public perceives that decisions are made without transparency, even diligent and responsible efforts may be viewed as predetermined outcomes.

Public attitudes further complicate governance. Many individuals differentiate between treating disease and altering inheritance, or between medical intervention and redesign. The same technology is perceived differently depending on its application. Governance systems originally designed for clinical trials and medical ethics are drawn into broader debates concerning intergenerational consent, disability, and inequality. Once these debates become prominent, opposition often solidifies rapidly, compelling researchers to defend the entire field rather than individual experiments.

Part of what drives that opposition is a recurring political pattern. Too often, those closest to the science end up deciding which futures are realistic, and therefore when broader democratic scrutiny is warranted. That can be done with good intentions. It can also leave the public largely reactive, stepping into the conversation only after major decisions feel underway. In gene editing, that dynamic showed up as the field moved from lab technique to clinical ambition and then to a global scandal. In SRM, it shows up when technical modeling, research agendas, and experiment proposals begin to shape what policymakers imagine as possible, before institutions have agreed on decision authority or public participation.

With SRM, the controversies around SCoPEx and the Alameda outdoor marine cloud brightening experiment went beyond physical risk to questions about decision making and consent.  What drove the reaction was process: who made the decision to proceed, who was consulted beforehand, and who only found out after the fact. In both cases, affected communities and outside observers felt excluded from decisions that carried implications for them. That feeling of exclusion, perhaps more than any technical concern, is what hardened opposition.

This issue is significant because SRM operates within social and media environments characterized by increasingly widespread mistrust and misinformation. The prevalence of the chemtrails conspiracy theory has already predisposed a segment of the public to view atmospheric interventions with suspicion. In these contexts, a failure in process not only damages the reputation of a single experiment but also reinforces existing suspicions about the entire field, thereby complicating future efforts at public engagement. Failures of legitimacy tend to spread more rapidly than subsequent corrections.

The field cannot wait for a crisis to build legitimacy after the fact. Governance structures need to be visible and participatory from the start, not introduced as a response to controversy. That means clear standards for who is consulted for different types of research, transparent communication about what is being studied and why, and accountability mechanisms that exist before they are needed. Public confidence is easier to maintain than it is to rebuild.

What SRM and gene editing share, and where they diverge

Gene editing and SRM share several features that make comparison useful. Both are framed as serving a public good, or at least preventing a public harm. Both intervene in complex systems where uncertainty is deep and surprises are likely. Both involve risks that are unevenly distributed, which makes coordination and consent politically difficult. Both are shaped by ethical disagreement within expert communities. And both exist in landscapes where fragmented governance can create incentives for boundary-pushing behavior. When standards are uneven, restraint can become a competitive disadvantage.

The differences, though, are not small. Human genome editing is deeply shaped by national cultural, political, and ethical contexts, which helps explain why regulatory approaches vary so widely. SAI, the form of SRM most likely to be deployed at scale, would produce immediate and shared planetary effects. Particles released into the stratosphere do not respect borders. The resulting changes in temperature, precipitation, and weather patterns would be felt globally, though unevenly, making politics inseparable from international relations and security. The mode of intervention also differs. Gene editing can be targeted to individual patients or specific organisms, while SAI would operate at the level of the climate system and would likely require ongoing management over time. Reversibility and monitoring differ too, and in complicated ways. Some gene edits are permanent. SAI deployment could be halted, but stopping could create abrupt rebound warming, which turns “reversible” into its own kind of trap.

Still, both technologies raise global human rights concerns and both challenge how societies govern interventions in systems many people view as shared inheritance. The United Nations Framework Convention on Climate Change frames climate change as a “common concern of humankind.” In the genome editing context, UNESCO’s Universal Declaration on the Human Genome and Human Rights describes the human genome, in a symbolic sense, as the “heritage of humanity.” Both frames point to the moral claim that some systems should not be governed solely by narrow interests, because the stakes spill across generations and borders.

SRM can derive practical lessons from the governance of gene drives. Gene drive governance has already established international forums and collaborative processes addressing oversight, risk, and transboundary implications. In contrast, the SRM field as yet lacks similarly robust and inclusive international governance mechanisms. A key lesson is evident: SAI governance has yet to achieve significant traction within international regulatory bodies, whereas gene drive governance has established venues under the Convention on Biological Diversity and its Cartagena Protocol on Biosafety. Given that SAI's effects would inherently cross borders and that unilateral deployment remains a possibility, developing comparable international governance infrastructure for SAI is an urgent, rather than theoretical, necessity.

Both gene editing and SRM illustrate the consequences of inadequate governance readiness. Institutions become reactive, controversies shape the field, and the public perceives the technology as advancing without consent. Once this perception is established, it is difficult to reverse. The policy challenge is not to devise perfect governance prior to any research, but to establish sufficiently legitimate structures early on. This approach ensures that research does not automatically lead to normalization and that boundary-setting is not determined solely by the most assertive actors.

That means clearer lines between stages, clearer expectations about participation, and governance that is practiced rather than only described. Otherwise the cycle repeats: capability surges, institutions scramble, and legitimacy fractures.