Key Takeaway:
Biotech companies like Colossal Biosciences are reviving extinct animals, such as dire wolves and woolly mammoths, by engineering them into modern organisms. However, the concept of de-extinction is not straightforward, as these creatures are not replicas but high-tech imitations. Modern science offers various approaches, including selective breeding, cloning, and synthetic biology. These techniques create modern organisms modified to carry ancient characteristics, but the exact fidelity to the past remains unknown. The language of de-extinction becomes problematic, as it evokes resurrection, but what is being done is more akin to remixing, combining fragments of old genetic information with the best available modern stand-ins. Gene editing holds immense potential, but it also comes with risks, such as entering ecosystems that have changed dramatically since their ancestors roamed.
In recent years, the dream of reviving extinct animals has leapt from science fiction to something approaching reality—or at least, something that looks close enough to convince us. With biotech companies like Colossal Biosciences at the forefront, the idea of walking among mammoths or witnessing dire wolves pacing across tundra no longer feels out of reach. But behind the hype lies a deeper question: are we truly resurrecting the past, or engineering something entirely new?
Take Colossal’s latest announcement: a litter of pups bearing key physical features of dire wolves, an apex predator that vanished over 10,000 years ago. The company has also launched high-profile efforts to “bring back” the woolly mammoth, the Tasmanian thylacine, and the dodo. These projects suggest that de-extinction is not just possible—it’s already underway.
But that depends on what we mean by de-extinction. Are these creatures truly the same as their ancestors, or just high-tech imitations?
The traditional understanding of de-extinction is straightforward: recreate an animal that once lived, using whatever biological means available. But modern science offers a patchwork of approaches. There’s selective breeding, where today’s animals are guided over generations to resemble their ancient cousins. There’s cloning, which uses preserved DNA to create a genetically identical organism. And there’s synthetic biology—engineering a genome piece by piece using tools like Crispr to insert traits from ancient species into modern relatives. Colossal’s work belongs in this last category.
The results of these techniques are not replicas, but proxies. They are modern organisms modified to carry ancient characteristics—like thicker fur, cold tolerance, or distinctive markings. That’s how Colossal approached the dire wolf: by editing just 20 specific genes in a gray wolf’s genome. The resulting animals may look like dire wolves, but their DNA is overwhelmingly modern.
A similar approach is being used to recreate the woolly mammoth—though what’s actually being built is a cold-adapted Asian elephant. Even after decades of research, scientists have only isolated a small portion of the mammoth’s genetic distinctions from its elephant cousins. Given the enormous complexity of an organism’s genome, this barely scratches the surface. It’s like trying to rebuild a cathedral with only a floor plan and a handful of original bricks.
That’s not to say these proxies don’t have value. A woolly-elephant hybrid might still serve an important ecological function in Arctic environments, helping to restore grasslands and slow permafrost melt. In this light, what matters isn’t fidelity to the past, but utility in the present.
The same logic is guiding Colossal’s efforts with the thylacine and dodo. In both cases, a close living relative—the fat-tailed dunnart for the thylacine, and the Nicobar pigeon for the dodo—is being used as the genetic chassis. Scientists are gradually swapping in ancient genes to replicate the appearance and traits of the extinct species. Whether this will result in an animal that behaves like its ancestor is still unknown.
This raises a philosophical question: at what point does a genetically modified organism stop being a modified version of something modern and start being a revival of something ancient? If 98.8% of our DNA is shared with chimpanzees, yet we are vastly different creatures, then how can a thylacine with 10% of its genome edited be considered a return of the original?
Meanwhile, efforts like the one underway to save the northern white rhinoceros offer a contrast. This is a true conservation initiative: using preserved cells and cloning to rebuild a population of an animal that still exists—barely. Unlike woolly mammoths or dodos, northern white rhinos have not been extinct for millennia, and the genetic material available is robust enough to aim for a full restoration, not a proxy.
This is where the language of de-extinction becomes problematic. It evokes resurrection—an exact duplication of a lost life. But what we’re actually doing is more akin to remixing: combining fragments of old genetic information with the best available modern stand-ins. The outcome may be useful or fascinating, but it is not a return. It is a reinvention.
And as powerful as this technology is, it also comes with risks. Edited animals will enter ecosystems that have changed dramatically since their ancestors roamed. Their health, behaviour, and ecological impacts are all unknowns. The risk isn’t just technical—it’s ethical. Are we building animals for nature, or for spectacle?
Still, gene editing holds immense potential. It could be used to strengthen endangered species, make wildlife more resistant to disease or heat, or even reduce the burden of harmful mutations. These are interventions that could prevent extinction, not just respond to it.
But to move forward responsibly, we need clarity. We must acknowledge that what is being created isn’t a mammoth or a dire wolf or a thylacine—it’s something new. A synthetic echo of the past, designed for today.
That doesn’t make it any less impressive. But it does mean we need a new vocabulary. These aren’t de-extinctions. They’re engineered adaptations, synthetic analogues, or ecological stand-ins. They’re not proof we can reverse time. They’re proof we’re learning how to shape what comes next.
