Jungwon
@jungofthewon
And the people bowed and prayed To the neon god they made
Every week @stuhlmueller kicks off our team sync with some ramblings about "Elicit at 30,000 feet." This week:
The idea of a fundamentally truth-finding AI system is noble. But the route is not to build a giant black-box general-purpose AI and then try to hit it with a stick until it tells "the truth." LLMs are powerful but their fundamental operation is not truth-seeking. I'm not sure…
Been working on this for a while, super excited to see it live! Near the start of the year, I committed to making Elicit’s data source access best in class, and after the Chrome Extension this is the second major public launch as part of that
Clinical Trials are live on Elicit 🎊 You can now search through the 545,000 trials available on clinicaltrials.gov and summarize them in structured tables or descriptive reports. New trial registrations are instantly available.
I've been thinking for a while that clinicaltrials.gov is an underutilized resource. Amazing data on what didn't work, but hard to analyze at scale Now Elicit runs the exact same systematic search > screening > data extraction > report flow we have for papers on trials!
Clinical Trials are live on Elicit 🎊 You can now search through the 545,000 trials available on clinicaltrials.gov and summarize them in structured tables or descriptive reports. New trial registrations are instantly available.
Clinical Trials are live on Elicit 🎊 You can now search through the 545,000 trials available on clinicaltrials.gov and summarize them in structured tables or descriptive reports. New trial registrations are instantly available.
i'm sure there's a lot to the founder acquihire deals that i'm not privy to... but this is my model of a founder's duty to their company
More people should create things to proliferate an aesthetic into our future, not just to solve problems. This is the quality that every artist and engineer I respect shares most universally. Without this, you are doomed to churning out slop.
improving our ability to model and price in risks from superintelligence
Insurance is an underrated way to unlock secure AI progress. Insurers are incentivized to truthfully quantify and track risks: if they overstate risks, they get outcompeted; if they understate risks, their payouts bankrupt them. 1/9
its not clear that LLMs need data to be structured in a different way. the reason why they are magical is because they can just absorb info as designed for humans. but definitely we should be able to move away from static, chunky artifacts and towards dynamic, easily composable…
I often rant about how 99% of attention is about to be LLM attention instead of human attention. What does a research paper look like for an LLM instead of a human? It’s definitely not a pdf. There is huge space for an extremely valuable “research app” that figures this out.
Today we introduce Elicit Alerts, a new way to stay informed about relevant research with AI.
Elicit powering Formation Bio's rapid R&D is the kind of AI impact we need in pharma. Love seeing tech cut through complexity and get life-changing treatments to patients faster! 🚀
“Elicit can extract technically complex variables from an enormous volume of papers with researcher-level accuracy – that’s incredible.” – Sandeep Basnet, MD, Senior Director of Clinical Development, Formation Bio Read how Formation Bio, a New York-based AI-native pharmaceutical…
so weird to live in a world where i have to be paranoid about my software subtly gaslighting me
We've added an estimated ~10 million more full-text papers to Elicit over the last month! This makes Elicit Reports more comprehensive/rigorous, since Reports will extract more info per paper.