Greg Eisenberg

And this whole microplastics thing is... Everyone's talking about it. And apparently they found that some huge percentage of men have microplastics in their testicles. So it went viral after that. In this article, I saw a link to one of the scientific journals, Nature or Science or just one of them. And I started reading one of the scientific papers and it was just so dense.

I mean, it's written for scientists. So hence my idea, that was the insight. The insight is there's tons of cool things going on in scientific papers, but they are written for scientists. What if we can extract, probably using AI, insights from scientific papers, curate it, and write it in plain English for people to read. I think that a lot of people would be down with that.

I mean, it's written for scientists. So hence my idea, that was the insight. The insight is there's tons of cool things going on in scientific papers, but they are written for scientists. What if we can extract, probably using AI, insights from scientific papers, curate it, and write it in plain English for people to read. I think that a lot of people would be down with that.

I mean, it's written for scientists. So hence my idea, that was the insight. The insight is there's tons of cool things going on in scientific papers, but they are written for scientists. What if we can extract, probably using AI, insights from scientific papers, curate it, and write it in plain English for people to read. I think that a lot of people would be down with that.

So I'll share my screen to give you a sense of how I think about these things and how I come up with some of these ideas. So I had that insight and then I went to perplexity Perplexity.ai, the AI search engine. And I was just like, what are the best places to find scientific journals and research papers?

So I'll share my screen to give you a sense of how I think about these things and how I come up with some of these ideas. So I had that insight and then I went to perplexity Perplexity.ai, the AI search engine. And I was just like, what are the best places to find scientific journals and research papers?

So I'll share my screen to give you a sense of how I think about these things and how I come up with some of these ideas. So I had that insight and then I went to perplexity Perplexity.ai, the AI search engine. And I was just like, what are the best places to find scientific journals and research papers?

Turns out there's Google Scholar, which indexes 200 million scholarly articles across disciplines. There's Base, which indexes 136 million academic documents. There's Core, which is 136 million. And then there's Semantic Scholar, which has 40 million articles. So first of all, I read that and I was like, well, I didn't realize that there were that many academic articles.

Turns out there's Google Scholar, which indexes 200 million scholarly articles across disciplines. There's Base, which indexes 136 million academic documents. There's Core, which is 136 million. And then there's Semantic Scholar, which has 40 million articles. So first of all, I read that and I was like, well, I didn't realize that there were that many academic articles.

Turns out there's Google Scholar, which indexes 200 million scholarly articles across disciplines. There's Base, which indexes 136 million academic documents. There's Core, which is 136 million. And then there's Semantic Scholar, which has 40 million articles. So first of all, I read that and I was like, well, I didn't realize that there were that many academic articles.

Think about how many insights are in these. I mean, of course, some are less relevant to the average person than others, but I bet you there's tons of insights there. So I see there's also some repositories, there's also libraries. So I see this, and then I actually, I want to go check out the Nature website, and I look at it, and this is... music to my ears.

Think about how many insights are in these. I mean, of course, some are less relevant to the average person than others, but I bet you there's tons of insights there. So I see there's also some repositories, there's also libraries. So I see this, and then I actually, I want to go check out the Nature website, and I look at it, and this is... music to my ears.

Think about how many insights are in these. I mean, of course, some are less relevant to the average person than others, but I bet you there's tons of insights there. So I see there's also some repositories, there's also libraries. So I see this, and then I actually, I want to go check out the Nature website, and I look at it, and this is... music to my ears.

When I see a website that looks like this, it sort of looks like it's been built in the late 2000s. There's nothing better than this. In fact, I can't even click here. It's not even letting me click because this pop-up is up. It's just, it's pretty brutal. So I really like that. And To me, this is validation that you have a lot of people who are referencing these articles.

When I see a website that looks like this, it sort of looks like it's been built in the late 2000s. There's nothing better than this. In fact, I can't even click here. It's not even letting me click because this pop-up is up. It's just, it's pretty brutal. So I really like that. And To me, this is validation that you have a lot of people who are referencing these articles.

When I see a website that looks like this, it sort of looks like it's been built in the late 2000s. There's nothing better than this. In fact, I can't even click here. It's not even letting me click because this pop-up is up. It's just, it's pretty brutal. So I really like that. And To me, this is validation that you have a lot of people who are referencing these articles.

You have hundreds of millions of articles and that there's just an opportunity to create a repository made for non-scientists. So how would I do this? How would I actually go and create something like this if I wanted to go build a startup? So first I would come up with, you know me, I like a catchy name.

You have hundreds of millions of articles and that there's just an opportunity to create a repository made for non-scientists. So how would I do this? How would I actually go and create something like this if I wanted to go build a startup? So first I would come up with, you know me, I like a catchy name.

You have hundreds of millions of articles and that there's just an opportunity to create a repository made for non-scientists. So how would I do this? How would I actually go and create something like this if I wanted to go build a startup? So first I would come up with, you know me, I like a catchy name.

So I think about what is a catchy name that kind of makes fun of the idea that scientific journals are for brainiacs. And I sort of think about that. I use that as my prompt and I just write a bunch of names related to that. And then I would actually probably use ChatGPT, OpenAI, and some other AI products to start fine-tuning a way to unpack scientific articles with one or two key insights.