I prompted the “deep literature” mode of Edison Scientific to write a comprehensive review on the biology of a gene that I know well. Twenty five min later it sent a 17 page PDF. It was, as best as I can tell, flawless. Folks, we are not in Kansas anymore. Don’t at me - try it.
Fyodor Urnov, Professor of Molecular Therapeutics at Berkeley, tested Edison Scientific's deep literature mode by requesting a gene biology review and received a flawless 17-page PDF within 25 minutes
AI Judge changed title after evaluation, original title: "Fyodor Urnov, professor of molecular therapeutics at Berkeley MCB, prompted Edison Scientific's deep literature mode to generate a 17-page gene biology review PDF in 25 minutes and called the output flawless"
Lior Pachter and collaborators shared screenshots of the demonstration.
Many users praise Edison Scientific's AI tool for its deep research mode, precedent search, and ability to cite relevant papers with clear reasoning in gene reviews.
No Digg Deeper questions have been answered for this story yet.
Most Activity
This is known as the denial stage of grief.
Apparently we're not in Kansas anymore.

@UrnovFyodor @SGRodriques I would encourage you to take any element of that paper you want to see visually and pass it to this.
One prompt to a one shot biology animation:
https://github.com/HarleyCoops/Math-To-Manim
Alternative: send to me and I’ll run. DMs open
I prompted the “deep literature” mode of Edison Scientific to write a comprehensive review on the biology of a gene that I know well. Twenty five min later it sent a 17 page PDF. It was, as best as I can tell, flawless. Folks, we are not in Kansas anymore. Don’t at me - try it.

in fairness/belated citation, i (the left-hand tweet) was inspired by the right-hand tweet. it's a good description of the situation.
am i banned from arXiv ?

You have to do a literature search rather than a kosmos run. You can view the result in the web UI, if you click on a reference it will pop up. You can click on references & see the quality labels Edison's assigned.
It's been a while since I've used Edison because I ran out of credits, but that feature was one of the things I liked the most about it.

@antonioregalado “Whom did it cite” ❤️❤️❤️ All the right papers.

@lpachter This whole nonsense inspired me to make http://pith.science last week.
Machine reviewed, community curated.
A new kind of preprint and journal

@UrnovFyodor @antonioregalado Currently running this (and other deep research tools) in bulk over the genes of multiple species https://github.com/ai4curation/ai-gene-review

@UrnovFyodor Their deep research mode is great, I also enjoy reading the reasoning over the papers (something you don't see in the PDF!).

@UrnovFyodor Fantastic. I’m curious how would eg Claude do on that?

@draparente Where is that feature? 🙏

@BioAI_Pharma Very.

@UrnovFyodor no small advance over imagined references

@UrnovFyodor That’s cool! How accurate are citations and references?

@christiancooper @SGRodriques Ooooooh!!!! How about the structure-function table? https://drive.google.com/file/d/1rA0qVcPFY64XiO_EPIQM-is47MPx--Z4/view?usp=sharing

@UrnovFyodor May we see it?

@UrnovFyodor yes, but who did it cite?

@UrnovFyodor did it handle isoforms/splice variants or just canonical summary

@ben_golub @lpachter should’ve gotten insurance

@antonioregalado Hear, hear