Synthetic Biology and Gene Editing: What Each Position Is Protecting

In December 2023, the United States Food and Drug Administration approved Casgevy — the first therapy based on CRISPR-Cas9 gene editing to reach clinical use. The treatment works by editing patients' own bone marrow cells to reactivate fetal hemoglobin production, effectively curing sickle cell disease and beta-thalassemia. Victoria Gray, the first American patient to receive an earlier version of the therapy in 2019, remained free of the debilitating pain crises that had defined her life for more than a decade. By early 2025, physicians at Children's Hospital of Philadelphia had gone further: they used a personalized CRISPR base editing therapy to treat a baby with carbamoyl phosphate synthetase deficiency, a rare enzyme disorder that is otherwise uniformly fatal. The child survived. These were not incremental advances. They were, by any reasonable measure, miracles of applied science.

Five years earlier, a researcher named He Jiankui had used the same underlying tools to edit the germline of human embryos — changes that would be heritable, passed to all future descendants of the twin girls born from that experiment. He had done it, he said, to protect them from HIV. He had not disclosed the work to regulators or obtained meaningful informed consent. He was sentenced to three years in prison. The scientific community's condemnation was nearly unanimous — but the reasons varied. Some were appalled by the recklessness of the specific act. Others were appalled by the category of act: heritable editing, regardless of how carefully done, crosses a line that had not been crossed before. The distinction mattered, because Casgevy and He Jiankui used the same molecular scissors. What separated the miracle from the crime was not the technology. It was where it was pointed, and who decided.

What biomedical optimists and therapeutic advocates are protecting

The lives of people who are suffering now — and the recognition that deferring powerful therapies on precautionary grounds is itself a choice with moral costs that tend to be invisible because the people who bear them have no voice in the governance conversation. Sickle cell disease affects approximately 100,000 Americans, the majority of them Black, and roughly 8 million people worldwide. Beta-thalassemia is the most common inherited blood disorder in the world, concentrated in populations across South Asia, Southeast Asia, and the Mediterranean. Before Casgevy, the only curative option was a bone marrow transplant requiring a matched donor — difficult to find, expensive, immunologically risky — available to a small fraction of patients. CRISPR-based therapies are a single treatment rather than a lifetime of blood transfusions. They transform the prognosis of diseases that have historically fallen below the research and development priority threshold because their patients were poor, non-white, or both. Advocates for therapeutic development are protecting the recognition that the precautionary principle, applied asymmetrically, reliably protects the healthy and the privileged by slowing therapies that would most benefit the chronically ill and the marginalized. Kiran Musunuru, who led the team that treated the baby with CPS1 deficiency, has argued that the ethical framework for gene therapy must grapple seriously with the cost of inaction — not as an abstract utilitarian calculation, but as an acknowledgment that every year of regulatory delay represents real patients who did not receive a treatment that existed.

The logic of the therapeutic pipeline — and the argument that germline editing, done carefully and for severe hereditary conditions with no other treatment options, is a continuation of medicine's existing commitment to preventing suffering rather than a departure from it. Not all advocates in this space endorse He Jiankui's act. But some argue that the categorical prohibition on germline editing — treating it as a line that cannot be crossed regardless of indication, careful governance, or patient benefit — is itself a position that requires defending, not assumed. Proponents of carefully regulated germline research point to the existing practice of preimplantation genetic testing: selecting embryos in IVF cycles that do not carry severe genetic conditions is already routine, already legal, and does not generate comparable moral alarm. If selecting against a heritable condition is acceptable when done at the embryo selection stage, the argument goes, why is editing the gene that causes the condition in a wanted embryo categorically different? The National Academy of Medicine's 2020 report on heritable genome editing did not conclude that heritable editing should be permanently prohibited — it concluded that current technologies did not yet meet the precision standard required for clinical use, and proposed criteria under which a responsible path forward could eventually be established. Therapeutic advocates are protecting the space for that path to remain open.

What biosafety and biosecurity advocates are protecting

The recognition that synthetic biology creates a class of risk that is categorically different from most industrial hazards — because engineered biological agents can self-replicate, evolve, and spread — and that the same knowledge enabling medicine enables catastrophe. Kevin Esvelt, who directs the Sculpting Evolution group at MIT and who, importantly, helped develop gene drive technology before becoming one of the most vocal advocates for restricting information about it, has argued that synthetic biology represents a fundamental change in the biosecurity threat landscape. The horsepox virus — closely related to smallpox — was reconstructed by a Canadian research team in 2016 from mail-order DNA fragments, at a cost of approximately $100,000, without any specialized facilities beyond a standard virology laboratory. The research was published in full. Esvelt's concern is not that any specific actor will immediately weaponize this knowledge; it is that the base of people with the technical ability to attempt it is growing rapidly, that published research continuously lowers the technical barrier, and that in a world of eight billion people, the probability of catastrophic misuse approaches certainty over sufficiently long time horizons if the knowledge continues to spread without constraint. Biosecurity advocates are protecting against a risk that does not announce itself in advance — that looks like ordinary scientific progress until it does not.

The norm of scientific openness — and the painful recognition that it may need to be partially revised for a specific category of research where unrestricted publication creates information hazards with mass-casualty potential. The open publication of scientific findings is one of science's foundational commitments: reproducibility requires it, credit assignment requires it, and the collective accumulation of knowledge depends on it. Biosecurity advocates are not arguing against this norm in general. They are arguing that certain specific categories of research — primarily research that enhances the transmissibility or lethality of potential pandemic pathogens, and research that describes methods for synthesizing dangerous agents from commercially available components — may require a different publication framework: one in which findings are shared with relevant biosafety authorities before or instead of full public disclosure. The gain-of-function research controversy, which predates synthetic biology but has been sharpened by it, illustrates the tension: experiments that modify influenza strains to test pandemic potential generate both useful scientific knowledge and a published recipe for a more dangerous pathogen. Esvelt's SecureDNA project is a specific proposal in this direction: cryptographic screening of DNA synthesis orders to prevent the creation of dangerous sequences, without requiring a centralized registry of what is dangerous. What biosecurity advocates are protecting is the possibility of governing a technology whose risks grow as its capabilities do — before the governance framework becomes necessary in the wake of a catastrophe.

What ecological precaution advocates are protecting

The integrity of wild ecosystems against a new category of irreversible anthropogenic change — one that operates at the genetic level and cannot be recalled once released. A gene drive is a genetic system that spreads through a wild population faster than Mendelian inheritance would allow: it copies itself into the matching chromosome in every individual it infects, ensuring near-100% transmission instead of the 50% expected for normal genetic variants. First proposed formally by Austin Burt at Imperial College London in 2003, gene drives were made technically feasible by CRISPR — which provided the precise editing machinery the concept had always required but never had. Target Malaria, funded in part by the Gates Foundation, has been developing gene drives designed to suppress mosquito populations that carry malaria in sub-Saharan Africa — a disease that kills more than 600,000 people annually, the majority of them children under five. The therapeutic argument for gene drives is, in its humanitarian terms, as compelling as any argument in this collection. Ecological advocates are not dismissing it. They are protecting the recognition that releasing a self-propagating genetic modification into a wild population is a one-way door: unlike a field trial, it cannot be stopped once it begins to spread. The population genetics models that predict drive spread are well-validated in laboratory conditions and small cage experiments. What they cannot tell us is what happens to ecosystems when a mosquito species is suppressed or modified at continental scale — because no experiment of that scale has ever been run, and running it is, by definition, irreversible.

The precautionary principle in a form that takes genuinely seriously the asymmetry between the timescale of ecological understanding and the timescale of regulatory approval — and the recognition that the communities most exposed to both malaria and to potential ecological disruption have had limited voice in the governance discussions that will determine whether drives are released in their environments. Gene drive governance has been the subject of sustained attention from international bodies, including the Convention on Biological Diversity and the International Risk Governance Council, both of which have emphasized that current governance frameworks were not designed for a technology that is both transnational (genetic modifications do not respect borders) and potentially irreversible. The governance gap is not primarily technical — it is procedural and democratic. The communities in the Sahel, in West Africa, in Southeast Asia who would be most affected by both malaria and by drive release have institutional systems and knowledge frameworks that are poorly integrated into the regulatory processes being conducted in London and Cambridge. Ecological precaution advocates are protecting the principle that the people most exposed to a risk should have meaningful authority over whether it is taken — not merely consultation, but decision-making power. Target Malaria has invested significantly in community engagement in Mali, Burkina Faso, and Uganda, and ecologists on the project genuinely believe the work is being done carefully. What precaution advocates are protecting is the difference between engagement and consent.

What justice and governance advocates are protecting

The recognition that the ability to edit human heritable traits at scale — once accessible — will reproduce and deepen existing inequalities, because access to powerful reproductive technologies has always followed the distribution of wealth. Casgevy, the first approved CRISPR therapy, costs approximately $2.2 million per patient in the United States. That price reflects development costs and market logic, not the actual cost of the therapy's manufacture — but it illustrates the pattern. Disability rights advocates and bioethicists at the Center for Genetics and Society, including Marcy Darnovsky, argue that a world in which heritable genome editing is technically available but economically restricted to the wealthy is worse — not better — than a world in which it is prohibited. The concern is not primarily about individual treatments for severe genetic conditions; it is about the trajectory of the technology once it has been normalized for any use. A governance framework that permits germline editing for sickle cell disease creates the infrastructure, the clinical practice, and the social license for germline editing for complex traits — intelligence, height, disease predisposition — as those targets become technically tractable. Disability advocates are protecting against the predictable consequence: a society in which certain categories of human variation are progressively edited out of the population, not by force, but by the cumulative weight of individual choices made available only to those who can afford them.

The lives and standing of disabled people — and the argument that the expansion of heritable gene editing implicitly communicates that lives like theirs are less worth living, regardless of whether that is the intention of any individual who makes an editing decision. The philosopher and disability activist Adrienne Asch articulated what has come to be called the expressivist objection: genetic selection against disabilities is not morally equivalent to abortion for any reason, because it is a selection against a specific trait — against the kind of person the fetus would become. This expresses a social message about which kinds of human lives are worth bringing into the world. Disability communities have been making this argument in the context of prenatal testing and selective termination for decades; synthetic biology amplifies its stakes substantially. The Autistic Self Advocacy Network's 2019 submission to the National Academies on heritable genome editing argued that the framing of autism and many disabilities as "diseases to be eliminated" reflects a medical model of disability that is contested within disability communities — that the suffering associated with many conditions is, in significant part, a consequence of inaccessible and inhospitable social environments, not of the disability itself. Justice advocates are also protecting the Global South's position in this governance conversation: the primary risks of gene drive technology are concentrated in tropical regions, the primary benefits of synthetic biology are being captured by pharmaceutical companies headquartered in North America and Europe, and the governance frameworks being designed are primarily shaped by institutions in wealthy countries. Gabriela Arguedas-Ramírez has called this dynamic "techno-scientific colonialist paternalism" — the pattern by which powerful biotechnology is governed by those who benefit from it, deployed in the territories of those who bear its risks, and framed as progress for all humanity.

What the argument is actually about

The synthetic biology debate is, at its core, a debate about whether the same tools can be safely governed differently for different purposes — and whether the moral logic that justifies therapeutic use can be kept from migrating into territory where it produces outcomes most of the involved parties would, on reflection, reject. The therapeutic case for CRISPR is genuinely compelling. So is the biosecurity concern. So is the ecological caution about gene drives. So is the justice critique of heritable editing. What makes this debate structurally difficult is that these concerns do not align into two teams — they cut across each other. Biosafety advocates and therapeutic advocates are not opponents; Kevin Esvelt and Jennifer Doudna share a laboratory tradition, and Esvelt has been explicit that his biosecurity concerns are not arguments against CRISPR therapeutics. Justice advocates and ecological precaution advocates are not the same community; disability rights objections to germline editing are about human lives, while gene drive concerns are about wild ecosystems, and they require different governance responses. The tendency to collapse this into a single "pro-synthetic biology / anti-synthetic biology" axis obscures the fact that most thoughtful participants in this debate accept some uses and oppose others — and the hard work is precisely in specifying which.

Whether the distinction between somatic editing (changes to an individual's own cells) and germline editing (heritable changes to all future descendants) is morally fundamental or merely technically convenient — and whether governance frameworks that permit the first and prohibit the second can hold as the technologies become more capable and less expensive. The current international consensus — expressed in the WHO advisory committee recommendations, the National Academies reports, and the policy frameworks of most high-income countries — treats somatic editing as acceptable in principle (subject to clinical evidence standards) and germline editing as not yet appropriate for clinical use (subject to future revision if criteria are met). This distinction has considerable moral logic: somatic editing affects only one consenting patient; germline editing affects all future descendants, none of whom can consent. But governance frameworks built on technical distinctions have a poor track record of stability when the underlying technology continues to advance. Base editing, prime editing, and other CRISPR variants are progressively improving the precision of germline modification. The price of DNA synthesis has fallen by roughly nine orders of magnitude since the 1970s. What is technically difficult today becomes technically routine in a decade. The question is whether governance frameworks designed for the current state of the technology will be robust to the state of the technology in 2040 — or whether they will require continuous renegotiation as capabilities advance, leaving the most consequential decisions to be made under the least favorable conditions.

Whether the Asilomar precedent — scientists voluntarily pausing recombinant DNA research in 1975 to allow governance to catch up — can be replicated for a technology that is simultaneously more powerful, more dispersed, more commercially developed, and being advanced by actors in dozens of countries with different regulatory philosophies. The 1975 Asilomar Conference is the canonical example of scientists exercising precautionary self-restraint before a governance crisis forced it. A group of molecular biologists, recognizing that recombinant DNA techniques raised safety questions they could not yet answer, voluntarily halted certain classes of experiments and convened a conference to develop guidelines. The moratorium worked, and the field emerged from it with a regulatory framework — NIH guidelines for recombinant DNA research — that held for decades. Whether this model is available for synthetic biology is genuinely uncertain. Asilomar worked because the relevant researchers were a small, relatively coherent community concentrated in a handful of American institutions, at a moment before the commercial biotech industry had developed significant interests in the field. Synthetic biology in 2026 involves university researchers, pharmaceutical companies, defense contractors, startups, iGEM student competitions, and government programs in dozens of countries. The community is not small. The interests are not aligned. And the technology has already migrated far enough from the research laboratory that a voluntary pause would be incomplete at best. What Asilomar offers is not a model to be replicated but a reminder that deliberate governance choices made before crises are more likely to hold than governance choices made in their wake.

Synthetic biology is unusual in this collection because the same tools that make it dangerous make it miraculous — and there is no version of the therapeutic promise that does not also carry the biosecurity risk, no gene drive that cures malaria that does not also demonstrate how to spread engineered traits through wild populations without consent. Most debates here involve a genuine conflict between values that can, at least in principle, be distributed to different institutions or different people. This one does not. The knowledge is the same. The choice is not between using it and not using it — that choice was made in 1972, in Paul Berg's Stanford laboratory. The choice now is about whether the governance of an irreversibly powerful technology can be designed before the worst applications of it are demonstrated, or whether, as has happened before with nuclear weapons and gain-of-function research, the governance arrives only after something has gone wrong. The answer to that question is not in the technology. It is in whether the communities with the most to gain, the most to lose, and the least institutional voice all end up in the room.