Hello everyone, my name is Mordechai Rorvig—I'm a writer, science journalist, and ex-physicist who was a regular participant on Less Wrong in the early days of the forum. I still occasionally check back here every now and then, and have always found lots of interesting writing here, but have mainly just been an infrequent lurker over the last decade. 

The purpose of my post today is to share a book project that I've been working on in the last year and which I'm now trying to raise funds to finish. The idea of the book is to investigate the remarkable but little-known evidence coming from neuroscience research, over the last decade, for interpreting modern AI programs like ChatGPT as closely related to simulations of brain regions, thereby giving a critical way to understand what they are, their strengths and weaknesses, how we should regulate them, make them safe for the future, and so on. It's a project that grew out of reporting I started doing in 2021, which was itself inspired partly by an important blog post by the user Gwern on the idea he called the scaling hypothesis.

Unfortunately, there's not a lot of funding for science journalists, especially when it comes to writing about science that is complex, recent, still a work in progress, and perhaps most of all, controversial in its implications. As a consequence, the idea that we are now commonly creating disembodied brain regions—in the form of AI programs—which a close reading of the neuroscience research might suggest, has hardly been talked about, at least, not in the media. I suspect it hasn't been discussed much here, either, on the forum—although I know neuroscience comes up here often. But the idea of the book is that this is potentially such an important and well-supported perspective that it deserve a serious, book-length investigation.

All this being the context, I launched a Kickstarter for the project on last Wednesday, with the goal of raising enough funds to work on the book full-time for another year, and then release the completed book in April 2026. You can support the project by purchasing a preorder for $10. You can find much more info about the ideas on the Kickstarter page, as well as 16 free sample chapters linked from it, which I have released to the public (for free) on a Substack.

I greatly admire the curiosity and expertise of the Less Wrong community, and I would be very interested to hear any thoughts, questions, comments, or objections on the project. It aims to be a work of popular science, by which I mean, something that makes the neuroscience research accessible to a broad audience, and easy to engage with.

Ultimately, I've worked on this project to try to make something that will be of benefit to the public, especially people like the users of Less Wrong, who are so interested in AI, and that's really been my main objective. That said, I'd obviously love to be able to keep working as a science journalist, as well, and so I would love to get your support, or any assistance in helping me get the word out, if you like the project. There are around 25 more days from today (Sunday, January 19) that I have to try and reach the funding objective. Thanks so much for your consideration.

Edit: Made a few small edits for readability and flow.