The question at the center of de-extinction and the Endangered Species Act sounds like a law-school hypothetical, but it now has a test case. In April 2025 Colossal Biosciences announced three white pups it called dire wolves, and within days the U.S. Secretary of the Interior was citing the technology in an argument about shrinking the endangered species list (Thomas 2025). Nobody has yet asked a court, or the U.S. Fish and Wildlife Service, whether such an animal is protected. This article works through the statute to see how far the text gets you, and where it runs out.
People clearly wonder about this. Headlines have asked whether a woolly mammoth needs a lawyer, The Wildlife Society ran a piece titled “Are dire wolves back and what would that mean?”, and political coverage tied the pups directly to the fate of the federal list (The New Republic). The legal answer matters for three practical reasons: who may legally harm or sell such an animal, whether a company could own it through a patent, and whether resurrected creatures could be used as a reason to relax protection for real wild populations.
This is general information, not legal advice.
The precise question
Strip away the spectacle and the issue has three parts. First, is a lab-made animal resembling an extinct form a “species” in the statute’s sense, and if so, which one? Second, if it belongs to a protected species, which of the Act’s prohibitions reach it, including in captivity? Third, does any of that interact with patent law, which decides whether the maker can claim exclusive rights?
Each part has a settled component and an unsettled one. The definitions are fixed text. How they apply to an organism nobody imagined in 1973 is not, and no agency rule, guidance document or judicial opinion addresses de-extinct organisms directly. That gap is the open question.

What the statute and regulations actually say
The definitions section of the Act, 16 U.S.C. § 1532, is broad by design. “Fish or wildlife” covers any member of the animal kingdom, including mammals, birds and invertebrates, along with their parts, products, eggs and offspring. Nothing in it asks how an animal came into existence. A wolf born from an edited embryo carried by a dog is still a member of the animal kingdom.
The word “species” is where things narrow. The statute says it includes any subspecies and any distinct population segment of a vertebrate species “which interbreeds when mature.” It never says what a species is at the base level. Carlin, Wurman and Zakim, in the most detailed law-review treatment of the topic, observe that the term itself is effectively undefined, leaving the call to agency science (Carlin et al. 2014). An “endangered species” is one in danger of extinction throughout all or a significant portion of its range, and “take” means to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture or collect, or to attempt any of these.
Listing and delisting run through 50 C.F.R. § 424.11. Paragraph (c) sets out five listing factors: habitat loss, overuse, disease or predation, inadequate existing regulation, and other natural or manmade threats. Paragraph (e)(1) provides that a species will be delisted when the best available data show it is extinct. The regulation does not define “extinct,” and it offers no route back onto the list for a taxon that later reappears through biotechnology, other than an ordinary new listing.
So much is fact. Everything below about resurrected animals is interpretation.
What the dire wolf case does and does not show
The factual record is thin and partly contested. Reporting by The Wildlife Society describes three pups, named Romulus, Remus and Khaleesi, born in October 2024 to a domestic dog surrogate after Colossal edited gray wolf embryos; the company said it changed 15 genes it associated with dire wolf traits, and the supporting genome analysis was a preprint (Learn 2025). Critics quoted in that piece describe the animals as white wolf-dogs, and one noted the company had not published which genes were altered.
The most authoritative taxonomic voice so far is the IUCN SSC Canid Specialist Group, whose task force issued a statement on 18 April 2025. Its verdict was blunt: the animals are not dire wolves and are not valid proxies under IUCN guidance, because there is no evidence they look or behave like dire wolves and because dire wolves and gray wolves differ by thousands of genes while the edits touched only a handful (IUCN SSC CSG 2025). The same statement warned that presenting gene editing as a fix for extinction could distract from protecting living species, including the gray wolf in the United States.
For legal purposes, this matters more than it first appears. Whatever one calls them, these animals descend from gray wolf cells. That single fact drives most of the analysis that follows.
Competing answers, steelmanned
Answer A: resurrected animals can be listed as the species they replicate
Carlin and colleagues make the strongest version of this case. Their reasoning is that the Act’s purpose is to conserve species and the ecosystems they depend on, nothing in the text excludes artificial origin, and genetic uniformity has not barred listing before; they point to the Franciscan manzanita, listed on the basis of a single known individual. Clones produced by somatic cell nuclear transfer from preserved tissue of a recently extinct animal are, on their reading, the clearest candidates. They conclude:
“In sum, the text and purpose of the ESA may be interpreted as qualifying resurrected species for listing …”
Carlin, Wurman & Zakim, 2014, 33 Stan. Envtl. L.J. 3, p. 30
Evidence for this view: the text is broad, listing factors would plainly be met by a handful of animals, and courts have accepted that hatchery-origin fish can count in listing evaluations, a precedent the authors cite. Evidence against: a species formally delisted as extinct has no ongoing status to revive, so the animals would need a fresh petition, and the agency would have to decide the taxonomic question first.
Answer B: engineered proxies are members of their source species
The same article draws a sharp line for genetically engineered facsimiles. An edited pigeon meant to resemble the passenger pigeon would, the authors suggest, most plausibly be treated as a distinct population segment of the living band-tailed pigeon, not as the extinct bird. Applied to Colossal’s pups, that logic, plus the IUCN verdict, points to one answer: they are Canis lupus with edits. The legal question then becomes the ordinary one of whether, and where, gray wolves are listed.
This view has the advantage of matching how taxonomists actually classify the animals. Its weakness is scale. If enough edits accumulate, at some point the proxy may diverge from the source species, and the statute offers no edit-count threshold.
Answer C: the Act was never meant to reach these animals
A third position holds that applying the ESA to de-extinction stretches it beyond recognition. A 2026 student note in the Arizona State Law Journal, summarizing Hope Babcock’s work, reports her view that the Act would likely apply but that doing so would run contrary to the statute’s expected application, and argues that governance should come through soft law instead (Woodward 2026). Carlin et al. also note that organisms created as pets or crops, such as GloFish, were never meant to be established in the wild, an analogy opponents can press. The strongest evidence for this view is legislative history: Congress was protecting animals on their way out, not animals on their way in. Against it, the text contains no origin requirement, and courts usually read the ESA broadly.
A doctrinal walk-through built for this article
To make the competing answers concrete, the table below applies the statutory and regulatory steps, one at a time, to three kinds of proxy animal. This is a structured thought experiment, not a prediction of any agency decision. Type 1 is a clone made from preserved cells of a recently extinct taxon. Type 2 is an edited proxy built from a living relative, the Colossal model. Type 3 is a clone of an individual from a species that is still alive and listed, which is closer to what Colossal says it did with red wolves.
| Step | Type 1: clone of extinct taxon | Type 2: edited proxy (e.g., the pups) | Type 3: clone of listed living species |
|---|---|---|---|
| “Fish or wildlife”? | Yes; origin irrelevant | Yes | Yes |
| Which taxon? | Most plausibly the extinct taxon itself (genetically identical to it) | Most plausibly the source species; IUCN group says not dire wolves | The listed species |
| Is that taxon on the list? | No; delisted as extinct under § 424.11(e)(1), or never listed | Depends on source species and location; gray wolf status varies by region | Yes |
| Route to protection | New listing petition; five factors likely met | Already covered wherever the source species is listed | Already covered |
| Take ban binds the maker? | Only after listing | Possibly now, if source species is listed where animals are held | Yes, subject to permits |
| Release mechanism | § 10(j) experimental population, likely nonessential | Same, as part of source species | § 10(j) or ordinary recovery |
| Patent eligibility (§ 101) | Weak: exact copy of a natural organism under Roslin | Stronger, if claims recite markedly different traits | Weak, same as Dolly |
Three results stand out. The first is counterintuitive: the edited proxies, the ones least like the extinct animal, may already sit inside the Act today, because their legal identity follows their source species. The Fish and Wildlife Service’s gray wolf page describes the species as endangered in much of the lower 48 states, with the Northern Rocky Mountain population and some other areas excluded, though the same page also carries an older narrative of delisting, so the precise current map should be checked against the Federal Register (USFWS, gray wolf species page). Colossal has not disclosed where the pups live. Whether captivity changes the result is a further question: Carlin et al. note that animals born in captivity before a species is listed are not subject to ESA permitting afterward, but that provision does not obviously cover animals produced after listing.
The second result concerns faithful clones. Here the extinct-species delisting rule is the obstacle. The listing would be new, and nothing in § 424.11 creates a fast track. Small numbers and genetic uniformity, often thought to be disqualifying, actually count toward listing under the five factors.
Third, patent law pulls in the opposite direction from species identity. In In re Roslin Institute (Edinburgh), decided by the Federal Circuit on May 8, 2014, Judge Dyk held that cloned livestock were not patent eligible. The opinion relied on Diamond v. Chakrabarty, 447 U.S. 303 (1980), and the Supreme Court’s 2013 Myriad decision, under which a natural organism is ineligible unless the claimed invention has markedly different characteristics from anything found in nature. The core sentence is short:
“Dolly’s genetic identity to her donor parent renders her unpatentable.”
In re Roslin Institute (Edinburgh), No. 2013-1407 (Fed. Cir. 2014)
The court rejected the argument that environmentally caused phenotypic differences saved the claims, because those differences were unclaimed, and it expressly left open that sharing nuclear DNA with a donor would not doom every clone claim. The implication for de-extinction is what this article calls a fidelity trade-off. The closer a resurrected animal is to the real extinct organism, the stronger its case for being that species under the ESA and the weaker its case for patent protection. The more engineered it is, the reverse. No court has tested this, and it rests on reading two separate doctrines together.
What the walk-through does not show: it cannot predict how the Service would classify any real animal, it assumes the cited law review reflects how courts would read the Act, and it ignores state wildlife law, the National Environmental Policy Act and genetically-engineered-organism rules, all of which Carlin et al. discuss and any of which might control in practice.

The political wrinkle: de-extinction as an argument against the list
The statute is not the only thing in play. When the pups were announced, Interior Secretary Doug Burgum wrote that the only thing his department would like to see go extinct is the need for an endangered species list, called the list a “Hotel California” where species never leave, and said the department was excited about de-extinction technology serving purposes beyond the recovery of lost species (Thomas 2025).
Legally, nothing in § 424.11 lets the existence of lab-made relatives count as recovery of a wild population. The five factors look at threats to a species in its range. A pen of edited pups in an undisclosed preserve does not change habitat loss for wild red wolves. Still, the remark shows why classification matters: if regulators ever treated proxies as interchangeable with the species they mimic, the existence of a backup could be offered as a reason to weaken protection for the original. The IUCN canid group raised a version of exactly this concern.
Where the law points, and what would settle it
What follows is opinion, labeled as such. My reading is that Answer B is the most likely outcome for anything resembling Colossal’s current animals: if the question were ever formally put, the Service would treat edited proxies as members of their source species, because that is what taxonomists say they are and because the alternative would require inventing a species from a press release. Answer A seems legally sound for true clones of recently extinct animals, but it would require a new listing and would likely be slow. Answer C is a policy argument, not a reading of the text, and it would likely lose in court.
Specific, checkable predictions and studies that would move the question:
- Prediction: no edited proxy will be listed under the ESA as the extinct taxon it imitates in the next decade. A listing rule naming such an animal as the dire wolf itself would falsify this.
- Prediction: if a listing petition or enforcement question ever arises over Colossal’s wolves, the Service will analyze them as Canis lupus. A written agency determination to the contrary would falsify it.
- Study: a peer-reviewed, published account of exactly which edits were made, and a formal taxonomic assessment of the resulting animals, would replace the current reliance on a preprint and press material.
- Policy test: a Fish and Wildlife Service notice or guidance on how genetically engineered or cloned animals relate to listed taxa would answer most of the table above in one stroke.
- Patent test: published patent applications claiming edited proxy animals by their altered traits would show whether the fidelity trade-off holds in examination practice.
Where this sits in the wider debate
The science behind proxies rests on genome editing, explained in our primer on how CRISPR works, and on the broader engineering mindset covered in what synthetic biology is building. The ownership half of the question has a cousin in patent law’s fights over who counts as an inventor, discussed in our piece on AI inventor rules. And there is a hard limit worth remembering: Carlin et al. state flatly that no method will bring back the dinosaurs, whose loss is covered in the asteroid evidence.

Common questions
Are Colossal’s dire wolves protected under the Endangered Species Act?
No agency or court has said. On the most plausible reading they are gray wolves with edits, so their status would follow gray wolf listings, which vary by region. The company has not disclosed where they live, so the answer cannot be pinned down from public information.
Could an extinct species be put back on the endangered list?
Only through a new listing. Federal rules delist a species once the best data show it is extinct, and nothing provides for automatic reinstatement. A clone population would need a petition, a taxonomic determination and a finding under the five listing factors.
Can a company patent a resurrected animal?
An exact clone of a natural animal is not patent eligible under the Federal Circuit’s Roslin decision about Dolly-style clones. An engineered animal whose claims recite markedly different traits stands on firmer ground. That is untested for de-extinction specifically.
Would the owner be allowed to sell or kill the animals?
If the animals belong to a listed species, the Act’s take and commerce prohibitions apply to any “person,” a term that includes corporations. Permits exist for scientific research and propagation. Without listing, ordinary state and federal animal laws would govern instead.
Further reading
- How to Clone a Mammoth: The Science of De-Extinction by Beth Shapiro. A scientist's book-length account of what de-extinction can and cannot produce, useful background for the species-identity question. See it on Amazon (affiliate link)
As an Amazon Associate I earn from qualifying purchases. A commission may be earned if you buy through this link, at no extra cost to you.
References
- 16 U.S.C. § 1532, Definitions (Endangered Species Act of 1973). Legal Information Institute, Cornell Law School. https://www.law.cornell.edu/uscode/text/16/1532
- 50 C.F.R. § 424.11, Factors for listing, delisting, or reclassifying species. Electronic Code of Federal Regulations. https://www.ecfr.gov/current/title-50/chapter-IV/subchapter-A/part-424/subpart-B/section-424.11
- Carlin, N. F., Wurman, I., & Zakim, T. (2014). How to permit your mammoth: Some legal implications of “de-extinction.” Stanford Environmental Law Journal, 33, 3. https://law.stanford.edu/wp-content/uploads/2018/05/carlin.pdf
- In re Roslin Institute (Edinburgh), No. 2013-1407 (Fed. Cir. May 8, 2014) (Dyk, J.). https://www.fr.com/wp-content/uploads/2016/11/In-Re-Roslin.pdf
- Woodward, M. (2026). In the era of Jurassic Park: Governing de-extinction through soft law. Arizona State Law Journal (student note; volume not stated in the posted PDF). https://arizonastatelawjournal.org/wp-content/uploads/2026/06/Woodward_PUB.pdf
- IUCN SSC Canid Specialist Group, Taxonomic Review Task Force. (2025, April 18). Conservation perspectives on gene editing in wild canids. https://canids.org/resources/CSG%20gene%20editing%20in%20wild%20canids.pdf
- Learn, J. R. (2025, September 23). Are dire wolves back and what would that mean? The Wildlife Society. https://wildlife.org/are-dire-wolves-back-and-what-would-that-mean/
- Thomas, M. (2025, April). After dire wolves are “resurrected,” Trump administration says U.S. endangered species list should “go extinct.” People, republished by AOL. https://www.aol.com/dire-wolves-resurrected-trump-administration-000333771.html
- U.S. Fish and Wildlife Service. Gray wolf (Canis lupus) species profile. https://www.fws.gov/species/gray-wolf-canis-lupus
- The New Republic. Why billionaire Trumpers love this dire wolf rubbish. https://newrepublic.com/article/193880/dire-wolf-extinction-trump
- National Geographic. Does a woolly mammoth need a lawyer? https://www.nationalgeographic.com/science/article/does-a-woolly-mammoth-need-a-lawyer
