Extinction is a part of nature. Of the five billion species that have existed on Earth, 99.9 per cent have vanished. The Late Devonian extinction, nearly four hundred million years ago, annihilated the jawless fish. The Triassic-Jurassic extinction, two hundred million years ago, finished off the crocodile-like phytosaur. Sixty-six million years ago, the end-Cretaceous extinction eliminated the Tyrannosaurus rex and the velociraptor; rapid climate change from an asteroid impact was the likely cause. The Neanderthals disappeared some forty thousand years ago. One day—whether from climate change, another asteroid, nuclear war, or something we can’t yet imagine—humans will probably be wiped out, too.
The difference with humans is that we’ve been taking a huge number of species down with us. Starting about three hundred thousand years ago, we learned to hunt with spears and in groups. That gave us significant agency in deciding which animals would disappear first—we chose them either because they wanted to eat us or because we wanted to eat them. The animals’ demise, though, helped doom large predators that hunted our preferred prey. Among the casualties were sabre-toothed cats and dire wolves. Along the way, various other species also breathed their last: woolly mammoths, Irish elk, dodos, carrier pigeons, Steller’s sea cows, great auks, thylacines (Tasmanian tigers). The carnage continues. Last year, the slender-billed curlew, a bird that once ranged over much of Europe and Asia, was declared gone. And there are only two northern white rhinos left—both females.
People have been sad about driving animals into oblivion for nearly as long as we have been eradicating them. And in recent centuries humans have tried to address the problem. In 1886, British authorities in South Africa were shocked by the speed with which Boer farmers had decimated the quagga, a half-striped relative of the zebra, and tried to save the species from extinction with the Better Preservation of Game Act. (The measure came too late.) In 1973, the U.S. Congress passed the Endangered Species Act, in response to the decline of many iconic American animals, including the bald eagle and the grizzly bear. Despite such laws and other conservation efforts, the current rate of extinction is, by some measures, a thousand times what it would be without humans.
Since the nineteen-eighties, various attempts have been made to see if it might be possible, somehow, to reverse the process. In theory, at least, the technological know-how that helped us extirpate so much wildlife could be deployed to bring back a few of our victims. Humans who are pursuing this goal are essentially asking for something that nature has never provided: a do-over.
Most of these investigations have been made by academic scientists or environmentalists. But what if the person trying to reverse an extinction was a man with an enormous amount of money, a mistrust of institutions, and a love of pop culture? The kind of guy who wants to move fast and fix things—but also increase his net worth. What animal would such a person choose to revive first? I saw the answer in late February, when someone turned on his computer and showed me a photograph of two cute white dire-wolf pups sitting on an asymmetrical throne made of iron swords. At first glance, it looked like an A.I.-generated image, but I was told that these were actual living animals. They were growing up at an undisclosed location, but in a few weeks he would let me go visit them.
Ben Lamm is a forty-three-year-old serial entrepreneur who has already had five “exits”—acquisitions of startups by other companies. He lives in Dallas; his estimated net worth is $3.7 billion. Lamm is dyslexic, and when he was younger he found reading difficult. He tended toward graphic novels and video games, but over time he taught himself, he says, to “read for concepts.” Today, he listens to a lot of podcasts devoted to bold new ideas. Among the interesting figures he has run across is George Church, a professor of genetics at Harvard Medical School. Church has endorsed using gene therapy to improve human resistance to radiation, thus facilitating interplanetary travel; he has also written about the possibility of cloning Neanderthals back into existence. A seventy-year-old with a long white beard and twinkling, wouldn’t-it-be-fun-to-try-that eyes, he is particularly admired among those who esteem speculative, gee-whiz thinkers; he has given half a dozen TED or TEDx talks. He likes to emphasize that he values the ways science can improve our lives. Church is also an accomplished lab scientist. He has more than a hundred and sixty approved or pending patents; among other things, he developed a process that allows crispr, the gene-editing technology, to be used to tinker with the human genome.
Soon after the pandemic hit, Lamm, who travels constantly, got covid, which led to a moment of introspection. He’d been founding companies since he was in college, and he’d made a lot of money, but had he made meaningful change? “Startups either grow and fail or grow and succeed,” he told me. Lamm had become interested in the way that various algae can capture carbon by absorbing it through photosynthesis before sinking to the bottom of the sea. Church had been part of a team that used Crispr to modify a blue-green algae’s genome so that it could sequester carbon twenty per cent more efficiently. Lamm called him and was excited by Church’s wide-ranging mind. “I’m really from the future, and I’m trying to let everyone catch up,” Lamm remembers Church joking. Lamm flew to Boston to visit him. Church talked about his myriad projects—he was using Crispr to help make pig organs suitable for transplants in humans, and to develop treatments for inherited diseases. Lamm found it all “pretty cool.” At some point, he asked Church which projects he would focus on if he had unlimited resources. Church said that one of the things he most wanted to do was bring back the woolly mammoth. Lamm was thrilled by the answer.
The woolly mammoth, which was native to the Arctic tundra, became extinct relatively recently, in evolutionary terms—about four thousand years ago. “The Egyptians were already building the Pyramids then,” Lamm told me. In 2012, Church explained, he had begun working with a nonprofit called Revive & Restore, in part to try to bring the giant creatures back to life. But it soon became apparent that the necessary technology was not sufficiently advanced. Lamm asked Church what kind of money he had access to for the project. Lamm remembers Church telling him that the tech investor Peter Thiel had donated to his lab, providing “a budget of a hundred.” Lamm was used to big money—“I’m not a scarcity guy, I’m an abundance guy,” he told me—but he was daunted. Fuck, I guess I could raise a hundred million dollars for this, too, he recalls thinking. Church then clarified that Thiel had spent only a hundred thousand dollars on the project. In startup terms, this was a pittance. Church hadn’t had the money to get the mammoth effort going for real, Lamm felt. Now he would.
In 2020, Lamm and Church agreed to create a for-profit company, called Colossal Biosciences, whose showcase product would be the deëxtinction of animals. “We are currently the apex predator,” Lamm told himself. “Why not use our technology for good?” (Jennifer Doudna, one of the inventors of Crispr, doesn’t see it this way. She believes that the editing technology should be reserved for essential matters, such as helping people with severe congenital disorders. As she wrote in the book “A Crack in Creation,” published in 2017, “If we can avoid altering nature more than we already have, shouldn’t we try to do so?”)
Lamm, who pitches as naturally as he talks, quickly raised sixteen million dollars from investors, promising them that Colossal would “spin the tech out” for a profit by, say, repurposing any advances that it achieved in genetic engineering for the benefit of human health. He rented space in Dallas and brought in about five million dollars’ worth of incubators and cell sorters. For embryology work, he installed a positive-pressure room, which prevents the inflow of possibly contaminated air from other areas. He assembled a team of computational biologists, cellular engineers, genetic engineers, and embryologists. Lamm would run the operation while Church did the deep thinking, the advising, and some of the lab work, along with lending the company his scientific prestige.
They made a shortlist of animals that Colossal would try to revive. The company would focus on what are known as “charismatic megafauna”—animals cute or scary or striking enough that their absence has left a significant mark on human consciousness. Lamm wouldn’t have been interested in starting on some tiny lizard or unloved beetle. The woolly mammoth was first on Colossal’s list. The next two were the dodo, whose extinction was facilitated by the arrival of Dutch sailors, who brought various invasive species on their ships, and the thylacine, a marsupial with the face of a fox and the stripes of a zebra. (Tasmanian farmers did not like the thylacine’s appetite for their animals, so the government paid for its slaughter; the last one died in a zoo in Hobart in 1936.)
Reviving an extinct mammal, Lamm told me, is essentially a high-tech challenge: you have to hack the problem with advanced tools. You need to collect DNA-fragment samples, typically from bones, and then synthesize the information from those samples in an effort to reconstruct the animal’s genome, at least partly. After that, he said, “we need to create the right media to grow the cells. We need to do immortalization of the cells”—modifying them to overcome their natural limit on division, thus allowing them to replicate indefinitely. Lamm became even more technical: “We need to create induced pluripotent stem cells and see if we can use a process called gametogenesis to create eggs and sperm of those cells. And then, if we are successful in that, we’d need also to consider how to optimize genetic diversity.” The ultimate goal, he emphasized, was not to create a one-off specimen but to have the species reoccupy its former habitat.
I asked Lamm whether Tom Chi, one of Colossal’s first investors and a venture capitalist known for his environmental awareness, was putting his money in to help save the planet. “He did not invest in us for fun,” Lamm said. “He invested in us because he thinks this company could be dual purpose—where it has a positive ecological benefit but it makes a fuck ton of money.”
In 1984, researchers detected traces of mitochondrial DNA in the skin of a taxidermied quagga in a museum. It suddenly seemed possible to bring back species that had died out. At the least, the road map was becoming clearer. Even reviving dinosaurs appeared to be within reach when, a few years later, papers reported the presence of DNA fragments in the remains of ancient reptiles that had gone extinct eighty million years ago. These fossils all turned out to be contaminated with modern DNA—in some cases, the researchers had sampled traces of their own genetic material. DNA degrades over time, and scientists now believe that a usable sample probably can’t be obtained from fossils much older than a million years. These constraints weren’t yet known, though, when Michael Crichton published his 1990 science-fiction novel, “Jurassic Park.” Lamm told me that, even now, one of the most frequent questions he gets is “Are you going to bring back the dinosaurs?”
In the nineties and two-thousands, researchers became better and better at extracting DNA, including from extremely small samples. In 2003, the human genome was sequenced—Church’s own DNA was the first to be publicly released. Numerous animal genomes followed. In 2012, Jennifer Doudna and Emmanuelle Charpentier developed Crispr, which relies on molecular machinery borrowed from bacteria to slice out genetic material from the nucleus of a cell and replace it with different genetic code. In 2014, the Times published its first article on the technology, declaring, “A genome can be edited, much as a writer might change words or fix spelling errors.”
Some scientists realized that Crispr could be used to insert a replica of an ancient DNA fragment into the cells of a closely related modern organism. The resulting hybrid creature might not be an exact copy of an extinct animal, but it could look very similar to the ancestor and thrive in the same ecosystem. Scientists soon found, however, that playing around with ancient DNA was harder than they expected. The samples they recovered were often damaged or impure; removing DNA from a fossilized bone sometimes damaged it further. Trying to accurately re-create an ancient genome involved looking at a modern cousin and estimating, by reverse engineering, which gene went where. Beth Shapiro, a noted ancient-DNA researcher who is now the chief science officer at Colossal, said, “We have to figure out how to build a trillion-piece puzzle while working with pieces that were left outside during a hurricane, using the picture of a slightly different puzzle on the top of the box, and the contents of more than a hundred and fifty thousand different puzzles inside”—that is, the DNA of all the microbes and fungi that got into the animal’s bone after it died.
In 2003, a group of European scientists tried to clone a recently extinct mammal called the bucardo, a mountain goat from the Pyrenees. The group used DNA that had been taken from the last living bucardo and then cryogenically frozen. They inserted the nuclei of bucardo cells into the eggs of fifty-seven goats, and then implanted them. Seven goats became pregnant; six had miscarriages, but one kid was born—with a malformed lung. Ten minutes later, that kid died, making the bucardo the only species to have gone extinct twice.
While struggling with the science, researchers were also contending with ethical objections to such projects. What if an extinct species’ original habitat was gone? Would you just be creating zoo animals? Was it immoral to alter an animal’s genome? Might concocted creatures suffer in unforeseen ways from the genetic changes? As with the bucardo, would you be allowing the same tragedy to occur a second time? In 2018, Shapiro, who was then running a lab at the University of California, Santa Cruz, made a striking remark to the Wall Street Journal: “There is no point in bringing the dodo back. Their eggs will be eaten the same way that made them go extinct the first time.”
Scientists such as Church were not daunted by the difficulty of the science or the ethical thickets involved. But in time they found a more persuasive framework for the deëxtinction concept which moved it out of the realm of fringe science. They emphasized that revived woolly mammoths, dodos, and the like would not just be curios. These creatures had once played valuable roles in nature, and their absence had left gaps. Returning the missing animals to their original ecosystems would help bring their habitats back into balance. Colossal, alert to this reorientation, focussed its ambition on species reintroduction.
Scientists at the company decided to take inspiration not from “Jurassic Park” but from recent “rewilding” efforts. One project that they often cited was the reintroduction of the gray wolf to Yellowstone National Park, in the nineteen-nineties. Headlines at the time had focussed on how angry the return of wolves had made local ranchers—gray wolves occasionally kill livestock, just as thylacines did. But, ecologists noted, the absence of wolves had left elk with few natural predators, and their population had exploded; their overgrazing had decimated vegetation that had sustained beavers and birds. Once the wolves were reintroduced, the number of elk declined and the park regained its equilibrium.
In interviews and at public events, Church emphasized that the reintroduction of the woolly mammoth could bring similar benefits to Siberia. By trampling shrubs, the beasts would encourage the growth of grasses that are good at absorbing greenhouse gases. In the winter, the mammoths, with their enormous weight, would tamp down snow, trapping methane—a greenhouse gas—that would otherwise be released by melting permafrost. Reviving the woolly mammoth, Church told CNBC in September, 2021, would bring Siberia “back to what, from a human standpoint at least, was a healthier ecosystem.”
Shapiro again played the skeptic. She opposed the idea of deëxtincting the woolly mammoth, arguing that creating one or two sample creatures would be cruel, given that they were likely highly social animals. She maintained that elephants do not do well in captivity or with assisted reproduction. And she insisted that elephants, instead of being impregnated with genetically edited embryos, “should be allowed to make more elephants.”
Shapiro is no longer Colossal’s gadfly. She started working with the company as an adviser in December, 2021, and became a full-time employee last year. Having pioneered techniques for the retrieval of ancient DNA and led a team at U.C. Santa Cruz that sequenced the dodo genome, she has a reputation in the world of genetics research that equals Church’s. But, if Church is deëxtinction’s id, Shapiro is its superego. She has expressed wariness, publicly, of projects that seem exploitative or merely flashy. Lamm began courting Shapiro soon after Colossal was founded. By that point, she had secured a lucrative appointment as an investigator at the Howard Hughes Medical Institute, so money wasn’t a strong lure. But she was impressed by how deeply he understood her work. He also caught her at a good time. She’d been instrumental in establishing the field of DNA recovery, and had won a MacArthur Fellowship, but, she told me, she was worried that her specialty had lost its Indiana Jones glamour and had become “routine—just another application of evolutionary biology.” She had noted, with approval, Church’s shift away from deëxtinction and toward rewilding, to creating “proxies for these extinct species . . . so they are useful in some way to the ecosystem.” In the spring of 2024, she signed on as the chief scientist for Colossal. “The money is about the same, but I’m trading tenure for stock options,” she joked to me.
In February, Shapiro joined Lamm and three other Colossal scientists in a nondescript twenty-five-thousand-square-foot lab in the Dallas Market area. Two more researchers joined the meeting on Zoom. Lamm divides his staff according to the animal they are trying to revive: Michael Abrams is one of the heads of the woolly-mammoth team; Sara Ord leads the thylacine group; Anna Keyte is in charge of the dodo project. Shapiro supervises them all, on both scientific and ethical matters. Lamm explained to me that he ran his operation more like a tech startup than like a typical lab: “There’s a healthy competition between the species leads.”
The woolly-mammoth group began the meeting. Abrams said that the team was about to publish a paper about some thirty mice that had been given genes to replicate the hair-growth patterns and cold-resistance capabilities of woolly mammoths. Trying out their research on mice was easier than doing so on elephants—for one thing, the average gestation period for a mouse is three weeks, whereas an elephant’s is twenty-two months, the longest of any mammal. Abrams said that the paper—which documented “a lot of really cool technology”—would be posted online in two weeks, on a site that makes research studies public before they are peer-reviewed; Colossal would try to explain its techniques to other biologists at the same time that its publicity team lofted them into the news cycle. Of course, the paper might not be accepted by a peer-reviewed journal. Lamm was clearly comfortable with the arrangement.
Abrams’s colleague Austin Bow, who was on the Zoom screen, said that the mammoth group had just got their hands on a promising new fibroblast line “derived from an Asian-elephant skin sample.” Fibroblast cells, which produce connective tissue, are crucial to gene editing and are easy to obtain. They can be stripped down to basic, flexible cells that can then be turned into any kind of cell in the body. A good fibroblast line maintains its genetic makeup over time even as it divides. The better the fibroblast line, the easier it would be to grow Asian-elephant cells that contained woolly-mammoth DNA and implant them in an embryo.
“Rocking the mammoth world here!” Shapiro said.
Ord, from the thylacine group, broke in with some news of her own. About a year earlier, the group, working in partnership with Australian researchers, had sequenced the animal’s DNA from a head preserved in ethanol at a museum in Melbourne. Colossal described it as the most complete sequencing ever done on an extinct nonhuman animal, mapping the thylacine’s genes with more than ninety-nine-per-cent accuracy. In October, 2024, Colossal announced that the group had successfully made three hundred edits to the DNA of a mouse-size marsupial called the fat-tailed dunnart, a living cousin of the thylacine with suction-cup ears and a ferocious appetite. The alterations would approximate aspects of the thylacine’s phenotype—the observable characteristics of a creature. The team had also discovered how to induce ovulation in the dunnart. The next step was to transfer a nucleus with the edited DNA into a dunnart ovum. But this was proving challenging. The mechanics of marsupial reproduction are not well studied. The exterior membrane of a dunnart ovum, which is about 0.2 millimetres in diameter, turns out to be very difficult to penetrate, Ord explained. The team had even used lasers to try to pierce the tissue. The ovum was so tough that they initially thought the laser was defective, but when they focussed the beam on a mouse ovum it “just blew up.”
That’s when, Ord continued, a colleague on the team with a background in mosquito genetics suggested trying to penetrate the dunnart ovum with a probe made of quartz, which is one of the hardest minerals in existence. The probe worked. Ord showed a video of it piercing the wall of the cell. Everyone applauded. Now they could take the nucleus out of the ovum, insert the nucleus containing the thylacine genes, and watch an embryo grow.
“Speaking of difficult challenges, what’s up with the birds?” Shapiro asked.
Keyte, the dodo director, reported that her group had been looking at the ninety thousand genetic differences between the dodo and its extant relatives, trying to map out which gene or genes were responsible for which characteristic. “I feel like a kid in a candy shop, because some of the genes that have come up on that are so cool,” she said enthusiastically. One bird whose DNA they had looked at was the Rodrigues solitaire—like the dodo, a flightless bird that had lived on an island east of Madagascar. The solitaire went extinct at the end of the eighteenth century, roughly a century after the dodo. The two birds have closely related genomes; both are distant relatives of pigeons. “You have a flighted bird that goes to an island, and it gets larger—it doesn’t need to fly anymore,” Keyte said. She had found a flight-related gene that had become inactive, giving the team a clue as to which of the genes in the dodo’s nearest living cousin, the Nicobar pigeon, they would need to alter to make Colossal’s avians earthbound, too.
Keyte added that her team was still a long way from bringing back the dodo. For one thing, the methods for growing and manipulating the embryonic precursors of avian sperm and eggs in a lab setting have been developed for only two birds: the chicken and, recently, the goose. Keyte said, “It’s been almost twenty years since culture conditions for the chicken were established, and those culture conditions have not worked for other bird species, even ones that are really closely related, like quail.” She added that, despite the dearth of related research, her team was getting better at growing the sperm-and-egg precursors in birds: “We’ve gotten to the point where we feel like we can start doing some migration assays”—a technique for studying how the cells in an early embryo begin to differentiate. Once the researchers got the basic method for growing bird cells down, they could use the technology not just to develop a dodo but also to help replenish populations of endangered birds. The team had already identified some species that could use the help.
“Awesome,” Shapiro said.
After the meeting ended, Lamm and I returned to his office. Colossal’s current lab space was just temporary—next door, the company was building a thirty-million-dollar state-of-the-art headquarters that is scheduled to open in June. Earlier, Lamm had walked me through the construction site, noting that the entrance would feature a ten-foot acrylic model of a woolly mammoth locked in simulated ice and shrouded in an emotive haze produced by a dry-ice machine. The model was being made by Wētā Workshop, the special-effects company that worked on the “Lord of the Rings” films. Peter Jackson, the director of that franchise, has invested several million dollars in Colossal.
In Lamm’s office, a wilted T. rex-shaped helium balloon peered out from a corner. I noticed a copy of Church’s book on deëxtinction, “Regenesis.” It “wasn’t easy going,” Lamm told me. He gave me a book he loved, “The Next 500 Years,” by Christopher E. Mason, a professor of genomics, physiology, and biophysics at Weill Cornell Medicine, in New York; it’s about how to biohack humans so that we can travel to distant planets.
Colossal’s roster of investors includes not only Jackson—who told Lamm that the company’s work is as fun as making movies—but also Paris Hilton. Image is very important to Lamm, who explained to me that his backers want to be associated with something exciting and sexy. He showed me some promotional videos that the company had made. These videos seemed to be pitched at the casually interested viewer—perhaps the sort of student he was as a young man—rather than at the ecological faithful. The voice-over in the videos, which nodded ironically to the steely voices of old-fashioned science documentaries, urged younger viewers to think of deëxtinction technology as cool. One clip ended with a voice grab from HBO’s “Game of Thrones”—“Here’s to the young wolf!” No one was going to confuse Colossal with the Sierra Club. The video, which showed the mechanics of deëxtinction as if they were starships hitting hyperdrive, was made by the genre-film director Michael Dougherty. “Mike did ‘Godzilla: King of the Monsters,’ ” Lamm said, with pride. “He did ‘Krampus.’ ”
This sort of hype has made Colossal an object of wariness in the scientific community, which prizes rigor and transparency. Hank Greely, a Stanford Law School professor with an expertise in bioscience, told me, “I understand why businesses often want to avoid disclosing as much as a peer-reviewed publication might require,” adding, “I know and like Beth Shapiro . . . and trust her quite a bit.” Still, he concluded, he would be happier to see their data available for other scientists to evaluate.
To build bridges with conservationists, Colossal has partnered with dozens of environmental organizations and promises to share with them any new techniques it develops, free of charge. It has created a foundation to distribute spinoff technology and given it fifty million dollars in funding. Bringing Shapiro on board was an important part of this outreach effort. (The company currently employs a hundred and thirty-two scientists—a cohort larger than many university biology faculties. It also funds forty postdocs at other institutions.) Lamm has won over other important figures. Love Dalén, a top woolly-mammoth researcher and a professor of evolutionary genomics in Sweden, was at first put off by Colossal’s swagger, but he told me he became convinced that Lamm and his startup were “dead serious at giving deëxtinction the best shot they can.” He is now on Colossal’s board of scientific advisers. The money that the company has raised, he said, “could of course have been better spent in protection of habitat, getting rid of invasive species, and combatting poaching.” But, he noted, “conservation funding is not a zero-sum game.” Lamm, Dalén felt, tapped funders who would not ordinarily give to, say, the World Wildlife Fund. The money invested in Colossal, he told me, would otherwise have gone to “some other tech companies, Bitcoin, the defense industry, or whatever.” And Colossal’s research is having valuable knock-on effects. Its work on woolly mammoths, Dalén said, is leading to a vaccine against elephant endotheliotropic herpesvirus, which, by some estimates, accounts for up to two-thirds of Asian-elephant deaths in captivity and a significant number in the wild. Ideas for the design of the vaccine, which is in field trials, came about through Colossal’s experience with mRNA manipulation and its research on the elephant genome. Mammoths appear to have been susceptible to the virus, too. (Lamm told me, with typical immodesty, “If successful, the vaccine will over time save more elephants than all of elephant conservation in history.”)
Lamm credits Elon Musk with stoking investors’ appetites to do “bigger things.” He adds, “I think people are, like, ‘Wow, we can do big, crazy things and there is a chance that we get a return on them.’ ” So far, Lamm’s investors have no reason to question their support. A month before I visited Colossal, the startup had completed its fourth round of funding. This time, it raised two hundred million dollars, adding to the two hundred and thirty-five million it already had. The company’s current valuation is $10.2 billion—on par with Moderna, the vaccine company. (One person who might miss out on getting rich from Colossal is Church, who, during the initial steps of founding the company, opted for a direct investment in his Harvard lab rather than equity. He has no regrets. Referring to Lamm, he told Forbes, “The fact that I’m not a billionaire is almost as interesting as Ben being one.”)
Colossal has already birthed three startups. The first, Form Bio, is developing machine-learning technology that can help pinpoint the optimal cuts that Crispr can make while editing genes. This would have applications to, say, the attempt to treat genetic conditions such as sickle-cell anemia or Huntington’s disease, and also to drug discovery and development. The second startup, Breaking, is making use of the advances Colossal has made in gene manipulation to reëngineer X-32, a microorganism that naturally degrades plastics, to increase its efficacy. “It takes plastics that have not broken down ever before and breaks them down in twenty-two months,” Lamm boasted. The third spinoff, he told me, was still “in stealth mode,” and he would not talk about it.
In truth, nearly everything Colossal is doing could find for-profit uses, since the attempt to genetically engineer, implant, gestate, and then birth an extinct animal touches on so many common health challenges. Currently, once the company has an embryo ready to grow, researchers insert it in a host animal, but such surrogates are limited in number and expensive to work with, and come with many potential health complications. Birth can also be traumatic—elephants rarely survive a Cesarean section, which is how many animals gestated in surrogates are born. So the company is working to replace the surrogates it would use—the Asian elephant for the woolly mammoth, for example—with artificial wombs. Lamm said, “We call it an exo-dev”—for “exogenous development”—“because it sounds less creepy.” (It didn’t.) Such technology could revolutionize human surrogate births, removing the ethical and emotional complications of the practice, not to mention the risks of pregnancy. “That field alone is a hundred-billion-dollar spinoff,” Lamm told me. An unaffiliated investment-advisory firm, Brownstone Research, has estimated that the “addressable market” for an exo-dev system for humans would be worth at least twenty billion dollars.
There is even money to be made in deëxtinction itself. While we were in Lamm’s office, he pulled up a slide titled “Rewilding: Monetization,” which showed a sketch of woolly mammoths in a chilly mountainous landscape. It was easy to imagine tourists gawking at the site.
“Jurassic Park!” I said.
“No, no, no, no, no,” Lamm responded. “That was an exaggerated zoo. This is letting the animals live in their natural habitats.” The distinction wasn’t entirely obvious to me, but I let him continue.
He explained that I was looking at a plan for a restored ecosystem. It was also a perfectly adapted money machine. There was a large area where the ancient elephants could graze, and this would be funded, in part, by carbon-offset payments from governments and corporations. The carbon value of a single elephant is about two million dollars, he told me. (An elephant increases biodiversity, in part, by spreading seeds in its dung and by crushing dense vegetation on forest floors, giving slow-growing trees the space to survive.) He added that the interesting educational opportunities and “sexiness factor” of Colossal’s creations would make its carbon credits “trade at a premium.” Rewilding could also provide value for nations that host revived animals. Shapiro had told me that when she visited Mauritius, where the dodo once lived, officials expressed enthusiasm about the prospect of increasing ecotourism. Lamm said, “If we are increasing the G.D.P. from tourism because of dodos, then we should partake in that.”
What about selling Colossal-created animals as pets? “We’re not in that business,” Lamm told me, adding, “I think you know us better than to even ask this.”
It was nearly time to see Colossal’s dire wolves. Since I was first shown the image of the two pups, I’d been told that a third had been born. She was equally cute. I’d also seen a photograph of one of the pups in the arms of George R. R. Martin, the author whose books were the basis of the “Game of Thrones” series, in which dire wolves feature prominently. “I used to watch the show, and I didn’t even think dire wolves were real,” Lamm told me. The entire dire-wolf operation had been done in stealth mode, and everyone involved had to sign nondisclosure agreements, including Martin.
Before making my travel plans, I spoke with Colossal’s chief animal officer, Matt James, who was low-key and modest compared with his voluble colleagues. “I just play with animals, mostly,” he explained. He said that he had recently been at an enclosure where Colossal had birthed four cloned red wolves—an animal that, despite being protected under the Endangered Species Act, was becoming rarer and rarer in the wild. As far as he knew, these pups were the first cloned red wolves ever. He added that, for the cloning process, Colossal scientists had used a new procedure that was less stressful for the genetic-donor animal. Usually, a biopsy is required to get clonable cells, but the scientists had figured out a way to obtain them merely by drawing blood. The new technique had been perfected by the dire-wolf team.
In “Game of Thrones,” the dire wolves are as big as lions. In reality, the dire wolf probably weighed about a hundred and fifty pounds, and it was only somewhat larger than today’s gray wolf. It also had more powerful jaws and a bigger head than a gray wolf. Dire wolves became extinct some twelve thousand years ago, at the end of the age of giant mammals. Most likely, their vanishing was not the result of direct contact with humans but, rather, was caused by the depletion of the large animals that were key to their diet. Dire wolves mostly ate horses and bison, with occasional forays into giant sloths and baby mammoths.
I was curious about how the dire wolf had slipped onto Colossal’s agenda, which was so clearly focussed on the three other extinct animals. Given Lamm’s alpha personality, it seemed possible that he simply wanted Colossal to dominate the deëxtinction space: it had begun with three animals, but why not four? Or perhaps the company, having made huge promises, felt the need to give concrete proof that it was making progress. According to James, the woolly mammoth’s due date had slipped from 2026 to 2028 after he pointed out the elephant’s long gestation period and other reproductive challenges. (Lamm says that Colossal is on track for 2028.)
Questions about Colossal’s timetables were evidently sensitive. During a tour of the lab with Lamm and the various species heads, I was escorted from workstation to workstation, so that I could better understand the high-tech genetic manipulations that would make the regenerated animals. It felt as if we were in the Central London Hatchery and Conditioning Centre in “Brave New World.” I asked Ord, the thylacine-team head, and Keyte, the dodo-team leader, when their animals would be resurrected.
“This decade,” Ord told me.
“The dodo is on track for this decade as well!” Keyte answered, competitively.
“No dates,” Lamm shot back.
As for the dire wolf, Lamm told me that the decision to put it at the front of the line was not because the company was impatient with its difficult marquee projects. In 2023, at a two-day conference with the team heads and Colossal’s advisory-board scientists, the attendees had whiteboarded what other animals might fit their guidelines for deëxtinction, and the dire wolf had generated a lot of enthusiasm. Lamm said that he had recently been visiting with people from Indigenous tribes, and many spoke of missing wolves, which had been central to their cultures. “There was this desire to bring back what they called the Great Wolf,” he claimed. Another reason—“and this is really important,” Lamm said—was that dire wolves are top-line talent in pop culture. They aren’t just in “Game of Thrones.” Dire wolves have starring roles in the video game World of Warcraft, the collectible-card game Magic: The Gathering, and the role-playing game Dungeons & Dragons. Lamm added that the Grateful Dead even have a song called “Dire Wolf.”
Colossal scientists began working on birthing a dire wolf in the summer of 2023. As I suspected, they were confident that the process would go faster than it was going with mammoths, dodos, and thylacines, in part because the genetics of dogs are among the best known of any nonhuman mammal. “It’s because people want to sequence their dogs and then talk about them,” Shapiro explained to me. Still, dogs and wolves split from each other some fifteen thousand years ago, and their genes have diverged over that time. “We had to develop a lot of resources for wolves, because they hadn’t been invented yet,” Shapiro said. “How to keep wolf cells healthy in culture, what to edit, et cetera. It’s still easier than a mammoth . . . but it’s orders of magnitude harder than a mouse.” Colossal hired experts in cell editing and wolf embryology and behavior.
The first problem the group encountered was where to get dire-wolf DNA good enough to use in sequencing the extinct animal’s genes. Shapiro happened to know as much about the topic as anyone in the world—in 2021, she had co-written a paper that traced the dire wolf’s lineage by using recovered DNA. While doing that research, she had examined the available dire-wolf fossils and found very little genetic material in them. But sampling techniques had grown more sensitive in the intervening years, and she decided to revisit the fossils. Two proved particularly useful. One was a piece of a dire-wolf tooth from Sheriden Cave, a late-Ice Age fossil pit in Ohio; the tooth was thought to be about thirteen thousand years old. The second candidate was a nearly complete skull from Idaho, found at the American Falls Reservoir. The skull was estimated to be more than seventy thousand years old, but contained an intact petrous—an inner-ear bone whose cells are densely packed, an arrangement that inhibits the entry of fungi and microorganisms that destroy DNA.
Because ancient-DNA reconstruction involves removing a small piece of bone, it puts delicate fossils at risk. Institutions aren’t always eager to oblige. But Shapiro had the clout to get the tooth shipped to her lab in Santa Cruz, and she sent one of her students to the museum in Idaho where the skull was kept. Nevertheless, she was nervous to begin drilling. Lamm reassured her: “Beth, you’re the No. 1 person in ancient DNA. I’m confident that you can do this. Like, you’ve done this a thousand times.” She used a dentist’s drill to make a small hole in the tooth and extracted powder containing billions of fragments of DNA. She then tagged each strand in such a way that a DNA-sequencing machine could read the order of the nucleotides—adenine (A), thymine (T), guanine (G), and cytosine (C)—well enough to provide a blueprint for engineering new cell lines. Lamm told me that the “Eureka moment” arrived when Colossal scientists realized that they would be able to establish much of the dire-wolf genome.(They ended up at ninety-one per cent.) Such efforts are never perfect. “There are errors in every genome that’s been assembled,” Shapiro pointed out.
The group then compared the DNA of the dire wolves with DNA sequences from gray-wolf cell lines. The gray wolf is the dire wolf’s closest living relative—they share 99.5 per cent of their DNA. If a gene does something in the gray wolf, the same gene in dire-wolf DNA likely codes for the same trait. The scientists’ goal was to alter the gray-wolf genome in ways that might make the resulting creature more like a dire wolf: principally, larger, with an increased head size and fluffy, light-colored fur.
The scientists could tell, based on the variation in key pigmentation genes, that dire wolves—like Ghost in “Game of Thrones”—had very pale coats. But the genes that guided coat color presented a problem: they carried with them a risk of blindness and deafness. (In humans, variations of these genes can lead to Waardenburg syndrome, which causes pigmentation deficiencies, among other problems.) So the group decided to edit a different gene that, when expressed in dogs, also codes for a lighter coat. This might make the dire wolf less authentic, but it would be better for the new creatures. “Otherwise, this animal is going to live out a suboptimal life because of choices you made,” Lamm pointed out. After almost a year of computational genetic analysis, Colossal researchers used Crispr to make twenty edits on fourteen genes. Fifteen edits were derived from Colossal’s study of the dire-wolf genome and five tweaks were derived from scrutiny of the gray-wolf genome. A Colossal-affiliated scientist spoke to me at length about which specific genes were altered, but the discussion was off the record, at the insistence of Lamm, who called such details the “company’s I.P.”
By mid-2024, Colossal researchers were watching thousands of edited cell lines grow in petri dishes stored inside incubators. Several cell lines stopped growing—perhaps because Crispr had mistakenly removed a bit of DNA or added code in the wrong place, or because the cell lines had other defects, or because they simply weren’t vigorous enough to keep dividing. By late summer, nuclei containing the edited genome were inserted into dog ova whose DNA had been removed. Forty-five embryos were placed inside two dogs. “Mutts, like a hound mix,” Lamm said. (Among other things, dog breeding is better understood than that of wolves.)
Only two embryos—one in each surrogate mother—grew to term. The gestation period was about sixty days, similar to that of gray wolves. The pregnant dogs were kept under supervision at a secure animal center and given weekly ultrasounds, starting about halfway through their term, to insure that the fetuses were growing normally. Lamm remembers “a chronic level of stress and anxiety knowing we were breaking new ground every day.”
On October 1, 2024, veterinarians—who had also signed N.D.A.s—delivered the pups, by Cesarean section. The team had hoped for vaginal births, James said, but as the weeks went on “there was a concern that there would be a size mismatch”—that the dire-wolf babies could be too big. When the pups were born, Lamm and Shapiro had just arrived in London for the première of a movie on ancient DNA. Lamm, whose first child had been born five months earlier, watched the dire-wolf births on FaceTime as James trained his phone on the operating table. The pups were large (one of them, James told me, was twice the normal weight of a gray wolf at birth) and white (gray wolves are born with very dark coats), with unusually robust heads. The pups were whining and crying. “I’m holding the first dire-wolf cubs in twelve thousand years,” James pronounced, adding, “It’s sort of surreal.”
For the first few days, the dire-wolf pups nursed on one of the dog moms, but, according to James, the mother “wasn’t able to keep up with their metabolic needs,” and they were switched to bottles. The two mutts were soon adopted by families, through the American Humane Society, without any mention of their role in lupine history. Lamm told me that he is very proud of how quickly the project came to fruition: “Within eighteen months of our putting the name ‘dire wolf’ down on a whiteboard, we birthed dire wolves!”
I asked Lamm when a university lab would have completed a comparable effort.
“Never,” he replied.
On March 25th, I went to meet the two older pups, named Romulus and Remus. (The third pup, named Khaleesi—for the light-haired protagonist of “Game of Thrones”—was not yet ready for visitors, I was told.) I was given directions to a one-acre enclosure at a location in the northern U.S. The dire wolves normally live on a two-thousand-acre property far from the spot where I was headed. “They live like kings there,” Lamm told me. As I drove up to the enclosure, I saw Church, looking like Gandalf, with a bushy white beard, and Lamm, who, with his luxuriant, glossy facial hair, resembles Jon Snow, Ghost’s master. Matt James was present, and a phalanx of animal-care officers stood in a corner, in case there were problems.
I suddenly saw two shocks of white—the young dire wolves. Both attentive and wary, the animals seemed not from this world. They were impressively large: roughly eighty pounds apiece, at five months old. Lamm told me that they would likely weigh about a hundred and forty pounds at maturity—at least twenty pounds more than a large adult male gray wolf.
James said, of the pups, “It’s hard to tell their girth, because they’re so hairy.”
Church recalled when he could almost hold a pup in each hand—he’d seen them twelve weeks after their birth.
Much of the dire wolves’ behavior reminded me of dogs’. Romulus and Remus rested on their haunches in the sun. They chased falling leaves; they chewed sticks. One peed, and the other hurried over to roll in it. Other aspects seemed wolflike—when Romulus got nervous, he did a sideways slide while facing us. (James explained to me that this maneuver is a way to both check out a threat and look as large as possible.) And, when the wolves ran, they loped as if their lower legs had an extra joint. They didn’t howl, and their footfalls were silent. Was this distinctive dire-wolf behavior? How could anyone say?
Though Romulus and Remus are identical twins—they came from the same engineered cell line—I could see that they were already behaving differently. Remus was braver. He would come up within ten feet of us, then think better of it. Romulus hung back. They both gave off a sense of biding their time; since nothing formidable had yet been asked of them, they had done nothing formidable. But they could; their bodies were clearly powerful. At one point, Remus seemed interested in getting behind us—he stopped when James looked at him—and I was reminded of the legend on the Dire Wolves card in Magic: The Gathering: “It’s amazing how scared a city kid can get at a dog.” I was told to keep a respectful distance. Lamm, though, seemed unafraid. “I pet Remus last week,” he said proudly. “But, you know, that’s because I’ve been around them forever.” (James says that around the time they enter puberty—at roughly a year old—they will likely be deemed too aggressive to be petted ever again. “We’re about to hit a trigger point,” he said.)
Church was eagerly looking for signs that the molecular engineering had done more than make Colossal’s wolves large and fluffy. The team had targeted an enhancer sequence that could produce the dire wolf’s pronounced muzzle and ears. James explained that cranial attributes take a while to develop. “I can hardly wait for the annual checkup,” Church said. “That’s where we’re going to learn, you know, where these phenotypes really distinguish them from the gray wolf.”
James said that Church would get his chance when the pups had full CT scans this fall, after turning one. “That’ll give us an opportunity to really look at musculoskeletal changes,” he said. Meanwhile, Lamm took more of a horse-breeder’s approach. The two wolves began to lope, then briefly run. “Watch! Watch the fur, watch the thickness,” he said. “You can see the musculature in the sun. Do you see it—shoulders? Side? Back? That’s not typical.”
“And the way the wind ruffles them,” Church added. “There we go. Watch the fur!”
Mike Dougherty had made a video for Colossal to proclaim the dire wolves’ birth. A voice declared, “We are pleased to announce that, for the first time ever in history, we have successfully resurrected a prehistoric pop-culture icon.” The dire wolves also got a boost from the Trump Administration. The Department of the Interior had sent Colossal a draft statement heralding the births as proof that “innovation—not regulation—had spawned American greatness.”
After seeing the wolves, though, I wondered how many people would accept that what I had just seen was a resurrection. Romulus and Remus embodied a lot of breakthroughs. No one else, Colossal said, had ever made so many precision edits to an animal genome at once. Nobody had ever birthed a mammal that had ancient DNA in it. “This is the first time these genes have been expressed in over ten thousand years,” Lamm pointed out. But to what extent were the pups dire wolves—as opposed to, say, gray wolves with some of the dire wolf’s physical traits? Did they act like dire wolves? No effort had been made to try to identify genes for behavior. Did they think or smell or sound like the extinct animals? “It’s not possible to bring back something exactly the way it was,” Shapiro told me, several times. That was a childish goal of the sort that had led her to challenge deëxtinction for so many years. She pointed out that, even within species, there was genetic variability, and environment and nongenetic factors alter living organisms. You had only to look at Romulus’s and Remus’s different personalities.
I spoke to Elinor Karlsson, a program director at the Broad Institute, a joint Harvard and M.I.T. facility. An expert on wolf and dog genetics, she is a member of Colossal’s advisory board. But she told me that she has some doubts about the company’s messaging: “I ask Beth, ‘Why are you calling this a dire wolf when it’s a gray wolf with seventeen or eighteen changes in its DNA?’ ” Shapiro has countered that she did what she set out to do: “We’ve succeeded in creating the phenotype of a dire wolf.”
This battle over definition had gone through a trial run just a few weeks earlier. Colossal had announced the existence of the “woolly mouse” that Abrams had spoken about at the species-lead meeting. The story was picked up everywhere—there were five billion media impressions in two weeks, Lamm told me. John Oliver and “Saturday Night Live” each played off it. (Colin Jost: “Now it’s on to step two—getting it drunk enough to have sex with an elephant.”) Mashable wrote that scientists had “accidentally created the cutest mice in the world.”
Science journalists were more cautious. They took issue with the fact that Colossal had used analogous genes from mice, not woolly-mammoth DNA, to achieve the effect of long, thick hair. A professor of evolutionary biology at the University of Buffalo did tell the Associated Press that the work was “technologically pretty cool.” But Nature noted that a Maine research facility has been offering its own long-haired mouse strain, named Wooly, for sale to scientists for the past twenty years.
Lamm was irritated when I mentioned the response to him. “We are the most advanced multicellular-synthetic-biology company on the planet,” he told me. The science behind the woolly mouse had been extraordinary, he reiterated—researchers had successfully made multiple gene edits on a living organism at the same time, a precursor to the quadruple-axel gene editing used in re-creating the dire wolf. “It was the most unique germ-line edit in any animal to date,” he said.
I asked Lamm if perhaps some overpromising by the Colossal brand had tamped down the applause. The word “de-extinction” appears nearly five hundred times on the company’s website; ordinary people could be excused for thinking that the word referred to creating an exact genetic replica of a once alive animal. Lamm responded, “I was warned when I started this business that some of the scientific community will be, if we are successful, jealous and somewhat frustrated.” He added, “You would think spending half a billion on deëxtinction and conservation would get them excited.”
Shapiro was more philosophical: “The pushback that we’re going to get from everybody is that it is not a hundred per cent of the way to the dire wolf. But we want to say with confidence that we’ve done this functional deëxtinction, and I think we have.” She added, “You know, as a scientist, would I have wanted to add another hundred million edits to the genome, to see what happened? Maybe—I mean, I don’t know.”
Church, for his part, is still hoping for a day when an entire extinct-mammal genome can be synthesized. Building a complete genome would likely be vastly harder than sequencing one and then synthesizing only small portions of it, but Church thinks that it’s technically possible. If you had the whole genome of an animal, and you inserted it into an egg, you would essentially be re-creating an individual from that extinct species. He has run the numbers. He told me, “The world has successfully synthesized a yeast genome, which is twelve million base pairs”—the fundamental units of DNA. The woolly mammoth had approximately three billion base pairs. “We’re only two-hundred-fiftyfold away!” he calculated. But Karlsson cautioned me, “We’re extremely far away from that goal now.” Lamm, not shy about big projects, said that filling in every gap in an extinct animal’s genome was “a cold-fusion-level problem.” He added that the difference between using a whole genome and using computational genetics to approximate the useful portions of a genome is “an academic distinction—which is probably why it interests George Church.”
Until the genome synthesis of mammals becomes a reality, Colossal’s dire wolf may be the closest we can come to the resurrection of charismatic megafauna. Lamm says that Romulus, Remus, and Khaleesi will not be allowed to breed; moreover, the company anticipates genetically engineering just three to five more of the animals. They will be kept on their preserve, protected by a ten-foot-high security fence and surveilled by drones—and visited, one suspects, only by the occasional billionaire. Colossal’s wolves are now at the age when their parents would teach them to hunt, but of course they have no parents; in fact, they have never seen other wolves. (There are some deer and squirrels on their two-thousand-acre plot, but that’s about it for other animals.) Perhaps a decade or so from now, when the wolves die—nature cares for nothing, all shall go—this genetic-engineering moon shot may, like the real moon shot, be remembered as much for the amazing new technologies it throws off as for the goal in itself. And the dire wolf may have the distinction, after the poor bucardo, of being the second species brought back to life only to die again. ♦
An earlier version of this article misstated the current stage of research on using Asian elephants as surrogates for woolly mammoths.
