Not Resurrection: What Can Really Be Brought Back from Extinct Species?

Twelve years after the idea seemed almost fanciful, gene editing, paleognomics and stem cells are already making it possible to create animals with some traits of extinct species. But Colossal's direwolves are genetically engineered gray wolves, woolly mice are not mammoths, and no extinct species has yet been reintroduced to the wild.

In 2014, an article was published on the Haydan website that called for serious discussion of the possibility Bringing extinct species back to lifeIn those days the use ofCRISPR לgene editing In its early days, ancient DNA reading was much more limited, and most of the discussion revolved around mammoths, Cloning And about the theoretical possibility of creating an animal that resembles an extinct species.

Twelve years later, it is no longer possible to be satisfied with the same discussion. Technologies have advanced greatly, private companies have raised hundreds of millions of dollars, researchers have assembled ancient genomes with a quality that was not possible at the time, and laboratories have created hairy mice and genetically edited gray wolves. But the gap between an engineering achievement and the title "bringing a species back from the dead" remains large.

As of September 2026, no extinct species has been reintroduced as a healthy, reproducing, and viable population. What science can do is clone an individual when suitable cells have been preserved, or create a genetically modified replacement from a living animal that carries some selected traits from an extinct species. This is an important achievement, but it does not erase extinction.

Three different ways hiding under the same name

the term "De-extinction"Connects several very different actions."

The first method is cloning by transferring the nucleus of a body cell into an egg whose nucleus has been removed. For this, a DNA sequence stored in a computer is not enough; a cell nucleus is needed that is preserved in a state that allows the embryonic development program to be restarted. In 2003, a cloned female Pyrenean ibex, a subspecies that had become extinct three years earlier, was born using this method. She died within minutes due to a health defect. This is the closest case to date of the birth of an animal from a species or subspecies that has already become extinct, and it also demonstrates how cloning alone does not create a population.

The second way is the deliberate breeding of individuals of a living species to recreate an appearance or traits reminiscent of an extinct population. Even if the result is very similar in appearance, it remains a descendant of the living species from which it was selected.

The third approach, currently the focus of attention, is editing the genome of a living species that is close to the extinct species. The genomes are compared, genetic variants associated with desired traits are selected, modified in living cells, and an attempt is made to create an embryo from the cells. The possible result is not a copy of the ancient species but a transgenic animal—a genetic or functional “replacement.”

Dire Wolves: An Engineering Achievement, Not a Return from Extinction

In April 2025, the company announced Colossal Biosciences Because she brought it back to life The direwolf, also known as the dire wolf (Aenocyon dirus), which went extinct more than ten thousand years ago. The company presented three white cubs: Romulus, Remus and Khalisi.

According to the company, the researchers edited 20 sites in 14 genes in gray wolf cells. The nuclei of the edited cells were then transferred into the eggs of enucleated dogs, and the embryos were implanted into surrogate dogs. Some of the edits were designed to create long, light fur, a larger body, and other characteristics that the company linked to the dire wolf. No dire wolf chromosome was inserted into the cells, and the basic genome remained that of a gray wolf.

This distinction is especially important because superficial resemblance is not evolutionary identity. An ancient DNA study published in Nature in 2021 found that dire wolves belonged to an ancient lineage distinct from living dogs. Newer research in Cell Genomics from 2026, which also included researchers and executives from Colossal, reconstructed two high-quality palaeognomes and updated the picture: According to the preferred model, about two-thirds of the dire wolf’s lineage came from a sister lineage to the group that includes the gray wolf, coyote and doe, and about a third from an earlier lineage. The new study teaches a lot about the animal’s complex origins, but it is about palaeogenetics—not proving that the engineered pups are members of the extinct species.

The original announcement did not include a peer-reviewed paper fully describing the creation of the cubs. Scientific review articles published later described them as phenotypic replacements or direwolf-like wolves. Analysis in The Anatomical Record determined that the edits affected a tiny fraction of the differences between the animals and that there was no basis to identify the cubs as dire wolves. The IUCN Canine Specialist Group went further and determined that the three individuals were neither dire wolves nor surrogates that met the organization's principles for restoring ecological function.

This does not negate the achievement. Multi-site editing in dog cells, cell testing, and the creation of live puppies by nuclear transfer are real technological capabilities. The problem begins when the name of an extinct species replaces an accurate description of what was created.

From woolly mouse to elephant: The road to the mammoth is still long

In March 2025, Colossal introduced “woolly mice.” The researchers altered up to seven genes related to hair structure, color, and fat metabolism, resulting in mice with long, wavy, and light fur. The experiment showed that it was possible to perform several edits together and get a relatively predictable combination of traits.

but The study on mice Published as a preprint and not peer-reviewed. It also did not show that the mice were more resistant to cold, nor were they implanted with mammoth genes. The researchers altered mouse genes in ways already known to affect fur and fat. This is an experiment in an editing platform, not a miniature mammoth, and not proof that a genetically engineered Asian elephant could develop, be born, and live as an animal adapted to the Arctic steppe.

At the same time, real progress has been made in the study of mammoths themselves. In 2024, researchers showed that the skin of woolly mammoth dating back about 52 thousand years, not only DNA fragments were preserved, but also information about The three-dimensional organization of chromosomesIn 2025, RNA patterns were also deciphered from tissues of a 39-year-old mammoth. The findings enrich our knowledge of gene and cell activity, but they are molecular information—not living cells from which a mammoth could be cloned.

In June 2026, a peer-reviewed study was published in Nature Methods reporting on Generating induced pluripotent stem cells from an Asian elephantThe researchers had to overcome, among other things, growth control mechanisms unique to elephants. Such cells may in the future help in studying development, creating reproductive cells, and preserving elephants. However, between a cell line in the laboratory and a normal embryo, a pregnancy of almost two years, calving, and raising a herd, there is a long series of unresolved problems.

A genome is not a species.

Even a perfect genetic sequence, if we had it, would not be enough to restore a lost species on its own. Animal development also depends on epigenetic marks, in the cytoplasm and mitochondria of the egg, in the uterus, on the mother's hormones, in the microbiome, and on environmental conditions. Social animals also have a learned culture: food selection, migration, care for offspring, communication, and coping with dangers.

Above all, there is the question of population. A single individual, or even several nearly identical individuals, is not a functioning species. A population needs genetic variation, appropriate age and sex ratios, habitat, food, protection from disease, and the ability to reproduce over generations. If the factors that led to extinction still exist, producing the animal in a laboratory does not solve them.

Nor is an ecological role automatically assumed with fur color or skull shape. To argue that replacing a mammoth, a dune, or a dire wolf restores an ecosystem, one must show what exactly the species' role was, whether the role is still missing, whether the habitat is capable of supporting a population, and what the effects would be on species living there today. This is whyIUCN Principles for Creating Replacements for Endangered Species The focus is on conservation benefits, animal welfare, risk assessment, and long-term monitoring—not on the mere ability to produce an impressive animal.

The practical benefit may come before extinction.

The technologies developed in the name of de-extinction may be more useful when applied to species that still exist. Cell and tissue banks, assisted reproduction, cloning, and gene editing can restore lost genetic variants to a population, strengthen disease resistance, or enable reproduction when very few individuals remain.

An important example is the black-footed ferret, an endangered species in North America. Cells from a female named Willa, who died without offspring, have been preserved since 1988. Several females have been cloned from them, and in 2024 United States Fish and Wildlife Service report Because Antonia, one of the clones, gave birth to two healthy puppies that survived. Here, it is not claimed that an extinct species has returned: cloning is being used to restore genetic diversity to a living species that is being tried to save. This is a more modest goal, but its benefit to conservation is clear and measurable.

There is also a cost. Cloning and embryo transfer procedures still involve failures, miscarriages, and the use of surrogates. The larger, rarer, and more social a species is, the slower the experiment and the higher the potential cost to animal welfare. An Asian elephant, itself an endangered species, is not just a potential recipient of a genetically engineered embryo. Any mammoth program must also account for the elephants who will carry the pregnancies and the cubs who will grow up without a herd of mammoths to teach them how to live.

What has changed—and what hasn't?

Since 2014, almost the entire toolbox has changed. It is possible to read ancient DNA and RNA in greater detail, reconstruct some of the structure of chromosomes, edit several genes simultaneously, produce Stem cells They are cloning and cloning dogs using advanced methods. Therefore, the idea is no longer pure science fiction.

What hasn’t changed is the meaning of extinction. A species is an evolutionary lineage, population, development, behavior, and ecological relationships—not a list of traits that can be ticked off. Science is getting closer to being able to create sophisticated replacements for extinct species. It still doesn’t bring back the past, and the replacement shouldn’t be presented as if it were the original.

The most important test will not be whether the new animal looks good in a photograph, but whether the research will produce a healthy population, whether it has a justified ecological purpose, and whether the same technology will also help prevent the next extinction. Until then, the precise wording is not “we have brought a species back to life,” but “we have learned to engineer a living animal to carry some of the traits of a lost species.”

Questions and Answers

Are Colossal's direwolves really an extinct species?

No. These are gray wolves that have undergone 20 edits in 14 genes and were created by cloning edited cells. They may carry some external traits attributed to the dire wolf, but they are not a reconstruction of the genome, lineage, behavior, or population of the ancient species.

Is it possible to clone a mammoth from a frozen carcass?

Not with the fragmented DNA found so far. Cloning requires a preserved, functional cell nucleus, not just a genetic sequence. The route currently being explored is editing Asian elephant cells to create a replacement with mammoth traits, and that too is far from a healthy embryo and a reproducing population.

Why isn't it enough to create an animal that looks like the extinct species?

Appearance is only a small part of biology. A species includes thousands of genetic and developmental traits, learned behavior, individual variation, and ecological role. Similarity in fur, size, or skull does not prove identity with the ancestral species and does not guarantee similar ecological function.

Do these technologies still have value for nature conservation?

Yes. Cell preservation, cloning, stem cells, and gene editing can expand genetic diversity and help breed endangered species. The black-footed ferret is a practical example. The benefits are clearer when the goal is to save a living population and its habitat, rather than simply creating an individual that resembles a lost species.

The scientific article

• On the ancestry and evolution of the extinct dire wolf — Cell Genomics, 2026
• The dire wolf (Aenocyon dirus) resurrection that wasn't — The Anatomical Record, 2026
• Derivation of elephant induced pluripotent stem cells — Nature Methods, 2026
• Three-dimensional genome architecture persists in a 52,000-year-old woolly mammoth skin sample — Cell, 2024
• Ancient RNA expression profiles from the extinct woolly mammoth — Cell, 2026
• Multiplex-edited mice recapitulate woolly mammoth hair phenotypes — bioRxiv, preprint, 2025

More on the subject on the science website

• Forum – Bring the extinct species back to life! / George Church
• Clone mammoths to protect the climate?
• The strange story of Yakinaf the mammoth - maybe she will come back to life
• My Sheep Tribe – 20th Anniversary of the Birth of Dolly the Sheep
• Research: Clones don't die young

For the original publication: Opening the original publication