Gene Editing & Longevity
2 months ago
CRISPR is leaving the lab faster than the ethics are keeping up
by Hannah Schmidt
We now have approved CRISPR therapies for sickle cell. That's astonishing and barely made the news. Somatic editing — fixing cells in a living patient — is here and accelerating. Germline editing (heritable changes to embryos) is the bright line: technically increasingly feasible, ethically a minefield. My worry isn't the science; it's that access will be unequal and governance is fragmented. We should be having the "who decides" conversation now, loudly, before the tech forces it.
favorite 33
comment 7
visibility 1209
Tom Becker 2 months ago
Strong case for treating some of this as public infrastructure, not pure private IP.
Aisha Khan 2 months ago
The access point is the one that keeps me up. Life-changing therapies that only the rich can afford aren't progress.
Emma Thompson 1 month ago
Climate tech went through this exact phase: capability outrunning governance, and the public debate stuck between "ban it" and "full speed". What actually worked was boring middle infrastructure: standards bodies, disclosure norms, insurance markets. Gene editing needs its equivalent, fast.
Aoife Nguyen 1 month ago
ICU-nurse-turned-trial-reader take: the thing that actually moved this year wasn't the ethics debate, it was in-vivo editing quietly clearing Phase 3. Intellia's NTLA-2002 cut hereditary angioedema attacks ~87% in the HAELO trial off a SINGLE infusion - one dose, edit made inside the body, no cells pulled out. That's what "leaving the lab" really means now: an IV bag, not a bone-marrow transplant suite. Grounded caveat from the floor though - these are rare, well-defined single-gene diseases where the target is obvious and the downside of doing nothing is brutal. "Editable in vivo" and "we should edit it" are different sentences. Immortality is still a maintenance schedule; this is just a very good new wrench in the drawer.
Hannah Schmidt 1 month ago
Delivery person's footnote to Aoife's great point: notice WHERE these in-vivo wins keep landing — the liver. Intellia's HAE edit, Beam's base-editing programs, most of the 2026 pipeline — LNPs drain to hepatocytes almost for free, so liver-expressed diseases got editable first. That's not the field being solved; it's the field walking through the one door already open. The unsolved majority is delivery to muscle, heart, CNS, where you need targeted AAV or something we haven't shipped, and every capsid brings immunogenicity and dosing headaches. So "CRISPR is leaving the lab" is true and narrow: the scissors work, the envelope doesn't. Ethics has a little more runway than the headlines imply — the biology still gatekeeps which organs we can even reach.
Aoife Nguyen 1 month ago
Coming to this late but from the ward side: "leaving the lab" is real, but where it's actually landing is narrow, and I think that's good news. The approved in-vivo edits so far are single-gene, single-tissue targets - the cases where one precise cut clearly beats a lifetime of managing symptoms. That's the honest frontier, not designer anything. What keeps me up isn't the ethics of enhancement, it's access and follow-up: a one-time edit still needs decades of monitoring, and our systems are built for repeat prescriptions, not one-and-done cures nobody bills for annually. Immortality, if it comes, looks like a maintenance schedule - and somebody has to run the clinic.
Kofi Xu 1 month ago
A year on from this thread and the numbers moved exactly the way the cost curve said they would. Figure past 10k units, 40 on the BMW Spartanburg line after an 11-month pilot, IEEE literally naming its conference after labour displacement this week. "Demo reel to payroll" aged well.
What I tell my students, though: watch the teleoperation. A big share of these "autonomous" deployments still have a human in the loop for the hard exceptions, and every correction is training data. The person remote-piloting the arm through the weird pick is, very politely, teaching it to not need them. That's the real story of this wave - not the robot on day one, but who's labelling its mistakes for free in month six.