A Company Is Bringing Back Dire Wolves. Tell Me Again How AI Is Only Bad.

Colossal Biosciences just did something that sounds like it belongs in a movie. They brought back dire wolves. Or at least something close to them.

In October 2024, two dire wolf pups were born — Romulus and Remus. A third, Khaleesi, followed in January 2025. The company used ancient DNA from a 13,000-year-old tooth and a 72,000-year-old skull fragment, then made 20 precision gene edits to gray wolf embryos to recreate traits that haven't existed in over 12,000 years.

Now, I want to be honest here. Scientists have pushed back hard on this. Some say these are really just gray wolves with a few modifications, not true dire wolves. Colossal's own chief scientist acknowledged they're "gray wolves with 20 edits." That's a fair criticism. Dire wolves diverged from gray wolves millions of years ago, and 20 edits doesn't bridge that gap.

But here's what I keep coming back to.

Ben Lamm, the CEO, has been open about the fact that this work depends on AI. He's said that with AI and gene editing converging, what used to take decades now happens in months. This isn't a guy hiding the ball — he's straight up telling you that artificial intelligence is what makes de-extinction even thinkable.

And Colossal isn't stopping at wolves. They're working on woolly mammoths, with calves expected by 2028. They're rebuilding the thylacine — the Tasmanian tiger that went extinct in 1936. They're even working on the dodo. An animal that's been gone for 350 years. They built a frozen biovault in the UAE to store tissue samples from 10,000 species.

This is where I lose patience with the anti-AI crowd.

I see people online every day screaming about how AI is destroying the world. And look, I get that the AI companies themselves have acknowledged real risks. Nobody serious is pretending AGI doesn't raise hard questions. But when you look at what's actually happening right now — not hypotheticals, not thought experiments — AI is doing things that seemed impossible five years ago.

Google DeepMind's AlphaFold predicted the structure of virtually every known protein. Over two million researchers across 190 countries have used it. That won the Nobel Prize in Chemistry in 2024. We're talking about understanding diseases, developing new drugs, and figuring out antibiotic resistance at a pace that would've been science fiction a decade ago.

And now a company in Dallas is using AI to bring back animals that have been fossils for thousands of years.

So when someone tells me AI is only a threat, I have to ask — a threat compared to what? Compared to losing species forever? Compared to diseases we can't cure because protein folding takes too long? Compared to a future where we just watch biodiversity collapse and shrug?

I'd rather be an optimist about this. Not a naive optimist — I know there are risks and I take them seriously. But the version of the future where AI helps us reverse extinction and cure diseases sounds a lot better than the one where we ban the technology because it makes some people uncomfortable.

The people who are loudest about hating AI tend to understand it the least. That's not a knock — most people haven't had a reason to dig into what these tools actually do. But if you're going to have a strong opinion, at least look at the full picture. Look at AlphaFold. Look at what Colossal is doing. Then tell me AI has no upside.

I'm not saying it's all sunshine. But I am saying it's an incredible time to be alive, and I'd rather lean into that than run from it.

Sources: Colossal Biosciences, CNN, Google DeepMind, NobelPrize.org