So cool to see the focus on really mining the scientific literature from @EdisonSci — it’s where a ton of valuable information is waiting and having models conversant in it is a big deal 👏👏👏
We should have a class of drugs you can only sell to people over 70 —super aggro longevity stuff, muscle regeneration, etc - tolerate more possible bad side effects given you don’t have much time left for effects to manifest...
Agree completely with @kevinweil this loop between models and robotic labs works today and it’s going to be exciting the directions that scientists take autonomous labs. Feels inevitable that it will yield a great speed up in our...
Not true. LLMs can run experiments with access to an autonomous lab and this is where they'll make scientific breakthroughs in the near term. Running experiments is about logical experimental planning (controls, replicates, etc), data analysis, and cycling on that. ...
My take on issue w cloud lab biz model so far — I have some ideas to improve this coming up, but this is fundamental issue from my standpoint.
I used my talk at the JPM conference this year to deep dive on @Ginkgo's future direction - building autonomous labs that replace traditional manually-operated laboratories. Wherever you see a lab bench today you will see one of Ginkgo's autonomous labs...
What’s exciting now with autonomous labs is you can get the closer to the flexibility of the bench but automated. So you could design 200 different experiments to run overnight and read out for you in the morning. ...

Great article from @packyM and @evanbeard -- I agree with this distinction between automation and autonomy. We've been working on autonomy with our autonomous labs at @Ginkgo -- much needed to get scientists off the lab bench, more in...