The definitive guide to gene editing.
Menu
Home Learn News Ask the Atlas
Explore Technologies Diseases Treatments Clinical Trials Companies Scientists Genes Research Institutions
Beyond medicine Agriculture Ethics Investing World map
Learn & Tools Start here Glossary A–Z Compare technologies Timeline Lists & Rankings AI Agents ★ Saved API
About About us Methodology Data Sources Editorial Policy Contact Disclaimers

🧭 Guided View
New to genetics? We explain every term as you browse, in plain English. Same pages, with the help built in.

⚡ Expert View
You already know the biology. Just the content — clean and compact, no extra explanations. This is the default view.

Interface language
Light mode

Ethics · Human editing

Germline and heritable genome editing

Editing embryos, eggs or sperm would change every cell of a future person and be passed to their descendants — which is why it is treated as categorically different from treating a patient.

How this page is written The Atlas does not take a position on contested ethical questions. Each page sets out the strongest version of the arguments on each side, states what is scientifically settled and what is not, and summarises the law where it exists. Where we think a claim is factually wrong — as opposed to ethically contested — we say so and cite the evidence.

The question, plainly

All the gene editing being used in medicine today changes cells in one person's body and stops there. Germline editing would change an embryo, so every cell of the resulting person carries the change — and so would their children, and their children's children. That difference is why almost every country treats it as a separate question with a separate answer.

Heritable human genome editing modifies gametes or embryos, producing changes present in every cell and transmissible to descendants. It is distinguished from somatic editing in law, ethics and regulation almost universally. Scientific concerns include mosaicism, unintended on-target and off-target changes, the impossibility of meaningful consent from the person affected, and the absence of a way to observe long-term consequences before they are irreversible across generations.

The arguments

What proponents argue
  • For couples where both partners carry two copies of a recessive disease variant, no embryo can be free of it — preimplantation selection cannot help, and editing is the only route to an unaffected genetic child.
  • Preventing a devastating condition before birth may impose less suffering than treating it afterwards, if it can be done reliably.
  • A blanket prohibition may push work into jurisdictions with weaker oversight rather than stopping it.
  • Some argue that a research pathway with strict oversight is more responsible than a ban that leaves no legitimate route.
What critics argue
  • Consent is impossible: the person most affected does not exist yet and their descendants never will be asked.
  • Embryo editing is frequently mosaic and produces unintended on-target changes; the result cannot be fully verified before birth.
  • Almost every case cited as a justification can be addressed by preimplantation genetic testing or donor gametes, making the genuinely unavoidable cases vanishingly rare.
  • The line between preventing disease and selecting for preferred traits is not defensible in practice once the capability exists.
  • It would be available first to the wealthy, and heritable advantage compounds across generations in a way no other medical inequality does.

Where the science actually stands

There is broad international agreement that heritable human genome editing should not proceed clinically at present. The WHO expert advisory committee, the US National Academies with the UK Royal Society, and most national bodies have concluded that safety and efficacy criteria are not met and that governance is inadequate. Disagreement is about whether the prohibition should be permanent or conditional on future evidence — not about whether it applies now.

Law and regulation

JurisdictionPositionInstrument
United States No federal statutory ban, but appropriations riders prohibit the FDA from considering applications involving heritable modification, which makes clinical use unavailable in practice Appropriations rider (renewed annually)
United Kingdom Prohibited for reproduction; embryo research permitted under strict licence up to 14 days with no transfer to a uterus Human Fertilisation and Embryology Act
European Union Broadly prohibited; the Oviedo Convention bars heritable modification for states that have ratified it Oviedo Convention; national law
China Prohibited; criminal liability established following the 2018 case Criminal law amendment and biosecurity legislation
Global No binding international instrument; WHO governance framework is advisory WHO recommendations, 2021

Summarised for orientation, not as legal advice; law changes and national implementations differ. Verify current law in the jurisdiction concerned.

The 2018 case

In November 2018 He Jiankui announced that he had edited CCR5 in human embryos and that twin girls had been born. The condemnation was near-universal, and the reasons are worth stating precisely rather than as general outrage: there was no unmet medical need, since HIV transmission is preventable by established means; the consent process was found to be deficient; the editing was mosaic and did not reproduce the protective natural variant; and the changes are heritable. He was convicted of illegal medical practice in China and imprisoned for three years.

The episode did more to shape governance of this field than any argument that preceded it.

Unresolved questions

  • Whether any future safety evidence could be sufficient, given that the consequences appear across generations.
  • Who would legitimately decide — national regulators, international bodies, or affected communities.
  • Whether the distinction between preventing disease and enhancement can be held in practice.
  • How to respond to work carried out in jurisdictions that do not participate in any framework.

Sources

Connected in the Atlas

Every entry on this site is linked to the others it relates to. These connections are part of the record, not a search result.

Genes

CCR5

Scientists

He Jiankui