Loading
Old version — revision 1
This is a fixed snapshot of Governance of human genome editing, saved by Import as part of the initial corpus import. It is not edited and it is not updated; the article may have changed since.
Edit summary: Initial import of content/governance-of-genome-editing.md — the filesystem corpus, unchanged. Not an edit.
The patchwork of national law, professional norms, funding rules, and international soft law that determines what may be done to a human genome and by whom.
Governance of human genome editing refers to the assembled rules — statutes, licensing regimes, funding conditions, professional statements, and international declarations — that determine which genetic interventions in humans may be attempted, by whom, and under what oversight. There is no global authority and no binding international instrument of general application. What exists is a patchwork whose strongest layer is domestic criminal and medical law, whose most visible layer is professional self-regulation, and whose weakest point is anyone unwilling to be bound by either.
Four distinct mechanisms operate, and they are frequently confused with one another.
Product regulation governs Somatic gene therapy exactly as it governs any other medicine. In the United States a gene therapy is a biological product requiring a licence; in the European Union it is an advanced therapy medicinal product. Trials require authorisation, manufacturing requires inspection, and approved products such as Casgevy carry long-term follow-up obligations, in some cases running to fifteen years. This layer is mature and works about as well as drug regulation generally does.
Funding conditions shape what research happens without prohibiting anything. The US Dickey-Wicker Amendment, in force since 1996, bars federal money from research in which human embryos are created for research purposes or destroyed. Since embryo editing research destroys embryos, it proceeds in the United States only with private money — which means the work is unfunded rather than unlawful, and correspondingly less visible.
Prohibition is where heritable modification sits. Most countries that address the question at all forbid transferring an edited embryo to a womb, though the instruments vary from criminal statutes to clauses in assisted-reproduction acts drafted before editing was possible.
Professional norms — summit statements, academy reports, journal policies, institutional review boards — do most of the day-to-day work and have no legal force. Their power comes from controlling access to publication, funding, and employment, which is considerable inside the academy and nil outside it. Legal scholars have noted that the resulting body of policy was mostly assembled before genome editing existed, out of instruments written for cloning, embryo research, and assisted reproduction, and fits the technology only approximately.1
The Council of Europe's Convention on Human Rights and Biomedicine, opened for signature at Oviedo in 1997, is the only binding treaty that directly addresses the question. Article 13 permits interventions on the human genome only for preventive, diagnostic, or therapeutic purposes, and only where the aim is not to introduce a modification in the genome of any descendants. It binds the states that have ratified it — around thirty — and several major research nations, including the United Kingdom and Germany, are not among them.
UNESCO's Universal Declaration on the Human Genome and Human Rights, also from 1997, names germline intervention among practices that could be contrary to human dignity. It is a declaration, not a treaty, and creates no obligation.
The three international summits on human genome editing, held in 2015, 2018, and 2023, produced statements rather than rules. The 2020 report of an international commission convened by the US National Academies and the UK Royal Society went further, setting out preconditions and a narrow translational pathway for any future heritable use, on the explicit basis that no such use was then justified.2 Its criteria are now the reference point in most technical discussions of Human germline editing.
| Dimension | United States | United Kingdom | China |
|---|---|---|---|
| Embryo research | Lawful, no federal funding | Lawful under HFEA licence, 14-day limit | Lawful under guidelines, 14-day limit |
| Heritable clinical use | FDA barred from reviewing applications | Prohibited by statute | Criminal offence since 2021 |
| Instrument | Annual appropriations rider | Human Fertilisation and Embryology Act | Criminal Law Amendment (XI) |
| Penalty for the act itself | No specific offence | Criminal offence under the Act | Up to seven years in serious cases |
| Exception created | None | Mitochondrial donation, 2015 | None |
The American arrangement is unusually indirect. There is no federal statute banning heritable editing. Instead a rider attached to appropriations legislation each year since 2016 forbids the Food and Drug Administration from acknowledging an application involving a heritable modification of an embryo. Since clinical use without FDA authorisation would be unlawful marketing of an unapproved biologic, the effect is a prohibition — but one that expires annually and could lapse through inattention rather than decision.
The United Kingdom regulates through licensing. Embryo research requires a licence from the Human Fertilisation and Embryology Authority, transfer of a genetically altered embryo to a woman is unlawful, and the one exception was created by an act of Parliament following public consultation: the 2015 regulations permitting Mitochondrial replacement therapy, under which the first births were reported a decade later. Whatever one thinks of the outcome, it demonstrates that a democratic society can decide such questions deliberately rather than by default.
China's regime was reconstructed after 2018. Ministry guidelines that carried no penalty were replaced by a specific criminal offence covering implantation of gene-edited or cloned embryos, provisions in the Civil Code on research involving human genes and embryos, and tightened ethical review procedures for human-subjects research. The He Jiankui affair was prosecuted under a general statute on illegal medical practice because nothing more specific existed at the time.
Beyond these, a survey of national policies published in 2020 found that a large majority of the countries examined prohibited heritable modification in some form, though many did so through ambiguous instruments never tested in court, and many others have no policy at all.3
The World Health Organization convened an expert advisory committee after 2018 and published its output in 2021: a governance framework, a set of recommendations, and a position paper.4 The recommendations are institutional rather than substantive. They call for a registry of human genome-editing research, mechanisms for reporting illegitimate work including whistleblower protection, international collaboration on oversight of cross-border activity, attention to intellectual property as a barrier to access, and education of scientists in governance. The committee stated plainly that no country should permit clinical heritable editing at present.
What the WHO framework is forIt does not regulate anything. Its purpose is to give national regulators a common vocabulary and a shared set of expectations, so that a country writing its first genome-editing law does not have to invent one from nothing. Judged as a coordination device it is useful; judged as a constraint it is not one.
The registry has the same character. Entries are voluntary, and the actors most likely to matter are those least likely to register.
Every consequence that has actually befallen anyone for editing a human genome came from a domestic court applying an ordinary law. There is no international inspectorate, no reporting obligation between states, and no mechanism by which one country can act on conduct in another.
The structural problem is that the required capability is widely distributed. Assisted reproduction is a global industry of thousands of clinics operating under highly variable regulation; editing reagents are inexpensive and shipped commercially; the expertise is taught in graduate courses. Jurisdictional arbitrage is already routine in reproductive medicine, where patients travel for procedures unavailable at home, and the same route is available for anything a clinic is willing to attempt.
Professional self-regulation, the model inherited from the Asilomar Conference on Recombinant DNA, assumes that the relevant actors care about standing among peers. That assumption held for decades because the only people who could do the work were academics. Privately funded ventures announced in 2025 that they intended to pursue heritable editing outside the university system, which converts the question from one about norms into one about jurisdiction: which regulator, in which country, has authority over a company that never applies for a licence and never files for approval. Related enforcement gaps recur wherever the technology escapes institutional settings — in Biohacking and grinders and self-experimentation, in Gene doping, and in the environmental release questions raised by gene drives under the Convention on Biological Diversity.
The line between therapy and enhancement is not codified anywhere useful. Most statutes speak of "serious disease" without defining it, leaving the boundary to be drawn case by case by regulators — the substantive difficulty examined in Bioethics of enhancement.
Selection is barely governed at all. Polygenic embryo screening is commercially available in some jurisdictions with essentially no regulation, achieves some of what people imagine editing would achieve, and involves no genome modification, so it falls outside every instrument described above. If In vitro gametogenesis makes large numbers of embryos available, this gap becomes the important one.
Access is a governance failure in its own right. The 2023 summit devoted much of its attention to the fact that approved somatic therapies reach almost none of the people who need them, arguing that a governance regime concerned only with restraint, and not with delivery, has answered half the question. That argument connects directly to Access and inequality, and to the pricing dynamics that have made several licensed cures commercially unviable.
Nobody is responsible for edited people. No framework assigns the duty to monitor the health of a person born from an edited embryo, or to pay for that monitoring, or to reconcile it with their privacy. Three such people exist. The number of institutions with an obligation to them is zero.
paperIsasi, R., Kleiderman, E., Knoppers, B.M. "Editing policy to fit the genome?" Science, 2016.↩A short policy-forum piece arguing that the instruments predate the technology, rather than a systematic legal survey.
reportInternational Commission on the Clinical Use of Human Germline Genome Editing. Heritable Human Genome Editing. US National Academy of Medicine, National Academy of Sciences, and the Royal Society, 2020.↩Convened by academies rather than by a regulator, so its preconditions bind nobody and carry no licensing force.
paperBaylis, F., Darnovsky, M., Hasson, K., Krahn, T.M. "Human Germline and Heritable Genome Editing: The Global Policy Landscape." The CRISPR Journal, 2020. ↩
reportWorld Health Organization. Human Genome Editing: A Framework for Governance and Human Genome Editing: Recommendations. WHO, 2021. ↩