Is De-Extinction Possible? What Science Can (and Can't) Actually Bring Back
A grounded, 2026 look at the technology, the headline projects, and the policy fights reshaping one of biology's biggest questions
"Extinct" used to mean permanent. A woolly mammoth skeleton behind museum glass, a stuffed dodo in a display case, a grainy black-and-white clip of the last captive thylacine pacing its enclosure in 1936 — these were treated as endings, not pauses. That assumption is now being tested in laboratories in Dallas, Melbourne, and Boston, where scientists are combining ancient DNA, CRISPR gene editing, and cloning technology in an attempt to reverse some of the animal kingdom's most famous disappearances.
The honest answer to "can extinct species come back" is: not exactly, not yet, and probably never in the way most people picture it. But the science has moved from thought experiment to living, breathing animal faster than almost anyone expected — and in doing so, it is dragging conservation law, biodiversity policy, and even the basic definition of a "species" along with it. Here is where things actually stand, based on peer-reviewed research, official policy documents, and reporting through August 2026.
What "De-Extinction" Actually Means
When people hear "de-extinction," they usually picture something like a movie plot: an animal cloned from perfectly intact ancient DNA and grown back exactly as it once was. That version does not exist, and according to most geneticists, it cannot. DNA degrades steadily after death; even in the best-preserved permafrost specimens, it survives only as short, damaged fragments. Nobody has a complete, undamaged mammoth genome sitting in a freezer, and nobody has living dire wolf cells to clone.
What today's science actually produces is what conservation biologists call a proxy species: a living animal, almost always a close genetic relative of the extinct one, that has been bred or edited to resemble or ecologically function like it. The IUCN Species Survival Commission, which published the field's first formal guidance in 2016, defines de-extinction as the creation of a proxy that is functionally similar to an extinct species — explicitly not a faithful replica of it. That distinction, proxy versus replica, is arguably the single most important idea in this entire field, and it is the one most headlines skip.
There are three broad routes scientists currently use to build a proxy species.
Figure 1. The three technical routes used to build a de-extinction "proxy species" today. None recreates the original animal's exact genome (IUCN SSC, 2016; NCBI PMC6265789).
Selective back-breeding is the oldest route: breeders cross living relatives over many generations, selecting for traits that echo the extinct ancestor — the way Germany's Heck cattle and the Quagga Project's striped zebras were bred to resemble the aurochs and the quagga. No new DNA is introduced; it is closer to genetic reconstruction than to editing. Cloning, or somatic cell nuclear transfer, takes a nucleus from a preserved cell and places it inside a donor egg; it produced the field's first, very brief success in 2003, when a cloned Pyrenean ibex (a wild goat subspecies) was born and died of a lung defect within minutes. Genome editing is the newest and most publicized route: tools such as CRISPR are used to rewrite specific genes in a living relative so that it expresses ancestral traits, guided by ancient DNA recovered from bones, teeth, or preserved soft tissue.
The Five Projects Racing Toward an Extinct-Animal Comeback
One company dominates this field commercially. Colossal Biosciences, founded in Dallas in 2021 by entrepreneur Ben Lamm and paleogeneticist Beth Shapiro, has raised more than half a billion dollars and was reportedly in talks this past July for a new funding round that could value the company between $20 billion and $30 billion. Its "de-extinction portfolio" now spans six species. Here is where each stands as of August 2026.
The Dire Wolf — Born, and Immediately Controversial
In October 2024, three pups named Romulus, Remus, and Khaleesi were born at an undisclosed preserve after Colossal's scientists made 20 edits across 14 genes in gray wolf cells, guided by dire wolf DNA recovered from fossils between roughly 11,500 and 72,000 years old. The company announced the births publicly in April 2025 and called them the world's first successful de-extinction.
The scientific reaction was swift and largely skeptical. Evolutionary biologist Vincent Lynch of the University at Buffalo argued the animals are essentially gray wolves with a handful of cosmetic edits rather than a resurrected species, noting that dire and gray wolves likely differ across millions of genetic positions, not twenty. UCLA's Bradley Shaffer, University of Maine paleoecologist Jacquelyn Gill, and UC Davis's Paul Knoepfler raised similar objections, and several pointed out that Colossal's claims were released through press channels rather than peer review. Colossal's own chief scientist, Beth Shapiro, later described the animals in interviews with more caution — as gray wolves carrying a functional dire-wolf phenotype, language considerably softer than the company's original announcement.
The Woolly Mammoth — Still the Flagship, Still Not Born
Colossal's original and best-known project aims to engineer cold-tolerance traits — shaggy hair, small ears, extra fat storage, cold-adapted hemoglobin — into Asian elephant cells, then implant the resulting embryo into an elephant surrogate for a roughly 22-month pregnancy. In March 2025 the company revealed an intermediate proof of concept: 38 "woolly mice," ordinary lab mice edited to grow long, coarse, golden fur using mammoth gene variants. In November 2025, Colossal acquired ViaGen, an established animal-cloning company that has cloned more than 3,000 animals, to strengthen its reproductive pipeline.
The birth target has slipped before. Colossal originally aimed for a mammoth calf by 2027, then 2028. As of late August 2026, CEO Ben Lamm told TIME magazine that the timeline has moved again, without offering a new firm date — a useful reminder that de-extinction schedules announced in press releases have consistently proven more optimistic than the underlying reproductive biology allows.
The Thylacine — Furthest Along on the Genome
Working with the University of Melbourne's Thylacine Integrated Genetic Restoration Research (TIGRR) Lab, Colossal has reconstructed a thylacine, or Tasmanian tiger, genome estimated at more than 99.9 percent complete, reportedly the most accurate ancient genome assembled to date, using tissue from a 110-year-old preserved specimen. The plan involves editing the genome of the fat-tailed dunnart, a small carnivorous marsupial and the thylacine's closest living relative, and using marsupial reproductive technology — including a synthetic pouch the company is developing — rather than a conventional pregnancy. Reporting from mid-2026 describes the project as having reached mid-gestation embryo development in lab models: a meaningful technical milestone, though still well short of a living, breathing thylacine.
The Dodo — Assembling the Toolkit, Bird by Bird
Because birds cannot be cloned using mammalian techniques, Colossal's dodo project relies on primordial germ cells — the precursor cells that become eggs and sperm — taken from the Nicobar pigeon, the dodo's closest living relative. In September 2025, the company announced it had cultured pigeon primordial germ cells for the first time and created gene-edited chicken surrogates: birds engineered not to produce their own germ cells, so that edited Nicobar pigeon cells can be injected in their place, eventually hatching offspring that carry dodo-like traits. Colossal has also formed a Mauritius Dodo Advisory Committee to help plan any eventual rewilding on the dodo's native island.
The Bluebuck — The Newest, Earliest-Stage Addition
Announced in April 2026, the bluebuck is a South African antelope that disappeared around 1800, one of the first documented mammal extinctions of the historical era. Colossal says it has sequenced and mapped the bluebuck's DNA and compared it against living antelope relatives, but the project remains in its analysis phase, with no gene-edited embryos reported yet.
Figure 2. Public status of Colossal Biosciences' five active de-extinction projects, plus the northern white rhino genetic-rescue effort for comparison, as of August 2026.
"It's Not a Dire Wolf": Why Scientists Are Divided on What Counts
Much of the disagreement in this field is not really about the biology — it is about language. Colossal's dire wolf press materials described the pups as a resurrected species, using the extinct animal's scientific name, Aenocyon dirus, as though the two were interchangeable. Critics say that framing collapses an important distinction: the pups carry 20 edited positions out of roughly 19,000 genes, while dire wolves and gray wolves are thought to have diverged more than two million years ago. Sean Stankowski, an evolutionary biologist at the University of Sussex, has been openly uncomfortable with declaring the project a successful "resurrection," even while agreeing that species boundaries are genuinely fuzzy in evolutionary biology. Ben Novak, a scientist at the de-extinction nonprofit Revive & Restore who has spent years defending the field, generally agrees the correct term is functional de-extinction — a proxy that fills an old ecological role — not literal resurrection. That more careful language rarely survives contact with a headline.
It is also worth noting that de-extinction research keeps generating tools useful well beyond any single extinct species. Marsupial assisted-reproduction methods developed for the thylacine project are already being applied to endangered Tasmanian devils and northern quolls, and elephant reproductive research from the mammoth program has produced a vaccine candidate against a herpesvirus that kills young Asian elephants. Researchers writing in the journal Heredity in 2025 describe this spillover as a "de-extinction and conservation flywheel" — the technology built for one goal ends up funding and advancing a different one.
Genetic Rescue vs. De-Extinction: The Northern White Rhino Isn't What You Think
The northern white rhinoceros is frequently mentioned alongside de-extinction projects, but it belongs to a different category. The subspecies is not extinct — it is functionally extinct, meaning the only two survivors, Najin and her daughter Fatu, are both female and cannot reproduce naturally. The international BioRescue Consortium, led by Germany's Leibniz Institute for Zoo and Wildlife Research, has used IVF with eggs harvested from Fatu and sperm preserved from deceased males to produce 39 pure northern white rhino embryos as of early 2026, intended for transfer into southern white rhino surrogate mothers. As of April 2026, only six embryo transfer attempts had been made, and none had yet produced a lasting pregnancy — researchers point out that human IVF itself took more than 100 attempts before its first success, as a way of framing the setbacks as expected rather than alarming.
This is genetic rescue of a living subspecies through assisted reproduction, not the revival of a species with no surviving members. It uses overlapping reproductive science, but the underlying challenge is meaningfully different, and lumping it in with dire wolves or mammoths tends to blur an important line in public understanding.
The Law and Policy Trying to Catch Up
Global biodiversity policy has only recently started to engage with this technology directly. The IUCN's 2016 Guiding Principles remain the field's foundational reference document: cautious in tone, explicitly non-endorsing, and calling for extensive risk assessment, genetic diversity planning, and stakeholder dialogue before any proxy species is released into the wild. In October 2025, the IUCN went a step further, adopting its first-ever global policy on synthetic biology and nature conservation — the broader technical category that includes gene editing and de-extinction — after roughly two years of drafting and more than 800 rounds of member feedback. The policy does not endorse or oppose the technology; it sets up a case-by-case evaluation framework and asks the Convention on Biological Diversity (CBD) and CITES to consider it in their own deliberations.
Those CBD deliberations have themselves been contentious. At COP16, held in Cali in late 2024 and concluded in Rome in February 2025, negotiators discussed synthetic biology governance, but a December 2025 academic review published in Transnational Environmental Law concluded that the final decision weakened proposed risk-assessment and monitoring language in favor of provisions emphasizing the technology's potential benefits — a shift the authors describe as a "de-legalization" of what had been a more balanced regulatory conversation.
In the United States, there is still no dedicated federal law covering de-extinction specifically; it currently falls into the gaps between existing wildlife, animal-welfare, and biotechnology regulation. What has changed is how the technology is being invoked inside the broader Endangered Species Act debate. After the dire wolf announcement in April 2025, Interior Secretary Doug Burgum publicly argued that de-extinction technology could reduce the need for strict species protections, framing gene editing as a potential substitute for regulation. Conservation groups, including Defenders of Wildlife, pushed back strongly, arguing that a gene-edited proxy created decades later is no substitute for preventing extinction in the first place. In November 2025, the U.S. Fish and Wildlife Service proposed a package of Endangered Species Act regulation revisions, and a separate bill from Representative Bruce Westerman would change how critical habitat is designated and limit judicial review of certain permitting decisions; both remain under review rather than finalized as of this writing. Separately, and with far less controversy, Colossal announced in June 2026 that it is partnering with the Fish and Wildlife Service on a genomic biobanking vault for currently endangered — not extinct — species, including red wolves, Hawaiian honeycreepers, and desert pupfish.
The Ecological and Ethical Case Against
Critics of de-extinction generally raise four connected concerns. First, resource diversion: money, media attention, and scientific talent spent on reviving extinct animals could instead protect species that are still alive but declining. Second, moral hazard: if political leaders treat de-extinction as a safety net, it may reduce the urgency around preventing extinctions in the first place, which is precisely the argument conservation groups raised in response to Secretary Burgum's comments. Third, ecological mismatch: the habitats and pressures that caused an extinction often still exist or have changed further, so even a successful proxy may have nowhere appropriate to live — one conservationist called it upsetting that Colossal's dire wolves were created with no genuine place left for them in the modern American landscape. Fourth, a definitional concern: because current techniques typically alter a small fraction of an animal's genome, the resulting creature remains, by most genomic measures, overwhelmingly the living donor species rather than the extinct one.
What's Realistic — And What Isn't
Functional proxies for recently extinct species with well-preserved DNA — the thylacine, the dodo, potentially New Zealand's giant moa — look achievable within years to a decade, based on current progress. Deep-time species like the woolly mammoth face harder biological limits: elephant reproductive biology is slow and poorly understood, surrogate mothers are scarce and already endangered, and each failed pregnancy raises real animal-welfare questions. Full genetic identity to the original species is not achievable with current or foreseeable technology; only trait-level resemblance is. And announced timelines, as the mammoth project has shown twice now, should be read as aspirational marketing rather than firm scientific commitments.
Frequently Asked Questions
Has any extinct species actually been brought back?
Not in the strict genetic sense. The closest historical case is the 2003 cloning of a Pyrenean ibex, a wild goat subspecies that died within minutes of birth from a lung defect. Recent high-profile projects like Colossal's dire wolf pups are gene-edited relatives carrying a small number of ancestral traits, not genetic replicas of the extinct species.
Which extinct species is closest to being revived?
Based on public project updates, the thylacine has the most complete reconstructed genome of any current project, at over 99.9 percent, though a living animal still appears to be years away.
Is de-extinction legal?
There is no dedicated de-extinction law in the United States or internationally. Projects currently proceed under general wildlife, biosafety, and animal-welfare regulation, guided by the IUCN's non-binding 2016 principles and its 2025 synthetic biology policy.
Why do scientists and companies want to de-extinct animals at all?
The most commonly cited reasons are restoring lost ecological functions, such as large grazers or predators that once shaped ecosystems, and developing reproductive and gene-editing technology that can be redirected to help species that are endangered but not yet extinct. Commercial interest is also a factor for private companies in this space.
Could de-extinction research accidentally revive a dangerous pathogen?
No — de-extinction projects work with animal genomes reconstructed from fragmented ancient DNA, not with reviving dormant microorganisms. Research into ancient, permafrost-preserved microbes is a separate scientific field with its own, distinct safety protocols.
The Bottom Line
De-extinction in 2026 is real science producing real, living animals — but it is not the resurrection the marketing sometimes implies. What exists today are engineered proxies: gray wolves with dire-wolf traits, elephants edited toward mammoth cold tolerance, pigeons on the way to becoming dodo-like birds. Each project is a genuine technical achievement and a genuine act of interpretation, and the gap between those two things is where most of the public confusion sits.
For students of zoology and conservation biology, this is a useful case study in reading science news critically: check whether a claim comes from a press release or a peer-reviewed paper, look for what fraction of a genome was actually changed, and ask who benefits from the more dramatic version of the story. The underlying genetics, population biology, and conservation-translocation frameworks behind de-extinction are worth understanding on their own terms — they will likely matter for how endangered species are managed for decades to come, regardless of how the extinct ones turn out.
Sources & Further Reading
- IUCN SSC, Guiding Principles on Creating Proxies of Extinct Species for Conservation Benefit (2016)
- IUCN, IUCN agrees first global policy on synthetic biology (October 2025)
- Convention on Biological Diversity, COP16 outcomes on synthetic biology
- Science / AAAS, "Is the dire wolf back from the dead? Not exactly"
- C&EN (ACS), "Dire wolf debate raises concerns on scientific overhype"
- Washington Post, "Citing dire wolves, Trump team aims to cut endangered species protections"
- Defenders of Wildlife, "Genetic De-Extinction"
- TIME, "The Woolly Mammoth's Comeback Is Taking Longer Than Expected" (August 2026)
- BioRescue / Ol Pejeta Conservancy, Northern White Rhino project updates
- Journal of Heredity, "De-extinction technology and its application to conservation" (2025)
- Bloomberg, "'De-Extinction' Startup Colossal Launches US Biobank for Endangered Species" (June 2026)
0 Comments