A New York City startup is storing frozen human embryos in a secret lab, editing their genes with AI-designed molecular tools, and openly pitching the world on the idea of creating genetically modified babies. This is legal, sort of. The editing part, anyway. Actually making the babies would be a federal crime. That distinction is, at the moment, doing a lot of heavy lifting.
The Lab, the Fog, and the Sales Pitch
NPR got an exclusive tour of Origin Genomics' undisclosed New York City location, because the founder asked them to keep the address quiet for security reasons. Which is a sentence that contains multitudes. Cathy Tie, Origin's CEO, suited up in a white lab coat, snapped on blue rubber gloves, and cracked open a cryogenic tank letting liquid nitrogen fog curl into the air like something out of a prestige sci-fi drama.
Inside those frosted glass tubes: human embryos. Donated by couples whose families carry hereditary diseases that have followed them for generations. Huntington's. Cystic fibrosis. Tay-Sachs. Hereditary cancers. These are not hypothetical horrors. These are real, devastating diseases that destroy real families, and Tie is not wrong to name them.
Origin's scientists extract stem cells from those embryos to test the company's experimental gene editors, according to NPR. The goal is to figure out which molecular tools can safely fix the defects. Tie is also testing AI-designed gene editors on other cell types carrying disease mutations, including cells with a BRCA mutation linked to breast cancer. Under a microscope, the cells glow fluorescent green when the editor makes it inside. "You're looking at the very first step," she told NPR's Rob Stein.
The Part Where We Mention It's Illegal
Using gene-edited embryos to actually create babies is against U.S. law. That fact is sitting right there in the NPR headline and Tie does not dispute it. What she disputes is whether it should be.
"What we are proposing is something completely revolutionary," she told NPR. "Apply this amazing precision genome editing to embryos that have a mutation before they are born. That allows the next generation to not be born with these diseases. It changes the entire paradigm of medicine as we know it."
That is an ambitious sentence. It is also the kind of sentence that sounds exactly like a VC pitch deck. Which, given the source of funding for companies like this, is probably not a coincidence.
We Have Seen This Movie Before
If the phrase "gene-edited babies" is making the hair on the back of your neck stand up, congratulations, you remember 2018. That was the year Chinese scientist He Jiankui announced he had used CRISPR to create the first genetically modified human babies, and the international scientific community reacted roughly the way you'd expect. He Jiankui was imprisoned for three years. The global consensus was that the technology was nowhere near safe enough, and that nobody had any business doing that yet.
The specific terror was off-target edits: the gene editor accidentally mutating something it wasn't supposed to touch, creating dangerous new mutations that could then be passed to every generation that followed. You do not get a do-over on germline editing. The changes go into the germ cells. They are heritable. Forever.
NPR reports that newer techniques like prime editing and base editing have meaningfully improved accuracy since then. Most scientists still think it's premature to try making gene-edited babies, and some think it may never be safe. But Origin Genomics and at least one other U.S. startup, Preventive in San Francisco, are pushing forward anyway. Silicon Valley venture capitalists and East Coast entrepreneurs are funding the charge.
The Ethics Are Not a Speed Bump
Tie's argument is a genuinely difficult one to dismiss. If you could spare a child from Huntington's disease, a condition that slowly destroys the brain and nervous system and kills everyone it touches, wouldn't you want to? The families who donated embryos to Origin are not abstract stakeholders. They are people who watched parents and siblings suffer and are terrified of passing it on.
But the history of medicine is littered with treatments that were sold on their benefits before anyone fully understood the risks. The specific danger here is not just individual harm, it's multigenerational harm. A mistake made in an embryo editor doesn't just affect one patient. It affects every descendant that person ever has. That is a categorically different kind of risk than almost anything else in medicine.
Some bioethicists do think the advances in editing precision are starting to shift the calculus, NPR notes. But "some scientists think we might eventually be close to maybe considering this" is a very different bar than "a startup CEO in a secret New York lab has frozen human embryos and a business plan."
The Dingo Take
You are supposed to look at this story and feel reassured by the white lab coats and the careful language about stem cells and induced pluripotent cells and "the very first step." You are supposed to note that no one is actually making gene-edited babies yet, that the law is still technically being respected, and that the intentions are good. Fine. Stipulated. But the architecture of what's being built here is unmistakable. A company. Frozen embryos. AI-designed gene editors. Venture capital. A CEO willing to say on tape that she wants to change the paradigm of medicine. These pieces do not assemble themselves into a research ethics seminar.
The He Jiankui story is instructive not just because he went to prison, but because of how it happened. He didn't announce a plan to eventually maybe edit babies someday pending regulatory clarity. He just did it. The scientific establishment was furious. The babies exist. The world moved on. The lesson a certain kind of entrepreneur might take from that story is not "don't do it." The lesson is "be ready to move faster than the regulations."
None of this means Cathy Tie is He Jiankui. Origin Genomics might be completely above board and genuinely committed to doing this responsibly, if it ever gets done at all. But "trust us, we're a startup" has not historically been the most robust ethical framework for decisions that affect the entire human germline. Someone in the federal government should probably be paying close attention to what's happening in that secret New York lab. Whether anyone currently employed by the federal government is paying attention to anything other than their own survival is, at this particular moment in American history, a genuinely open question.
Comments