Gord Fishell

Thoughts on Anthropic and the pentagon....The US military has 3000 strikes in Iran, all determined using Claude's help -assume 94% accuracy and 0 hallucinations. So 180 errors in targeting. How many civilians are killed in those errors? That is why we can't autonomize warfare!

Excited to be on the organization meeting for the 6th annual NCM conference on Neural Circuits in Health & Disease. Terrific line up of speakers!

Pyramidal neurons proportionately alter cortical interneuron subtypes - Nature Pyramidal neuron-driven mechanisms actively promote the survival and terminal differentiation of their associated interneuron subtypes.

Excited for the publication of our recent work on how pyramidal cells shape the identity of interneurons in the Cortex! Big shout out to @artofbiology and Sherry Jingjing wu! www.nature.com/articles/s41...

Biological fidelity: The engine driving the neuromorphic renaissance | PNAS Biological fidelity: The engine driving the neuromorphic renaissance

Thrilled to see our PNAS paper featured in @giacomoi.bsky.social commentary. A fantastic recognition of how biologically grounded cortical microcircuits can accelerate the next generation of NeuroAI—and a meaningful step connecting it to neuroscience www.pnas.org/doi/10.1073/...

LARIS enables accurate and efficient ligand and receptor interaction analysis in spatial transcriptomics Advances in spatially resolved transcriptomics provide unprecedented opportunities to characterise intercellular communication pathways. However, robust and computationally efficient incorporation of ...

Excited to report the publication of LARIS, a program that can predict spatiotranscriptic LIgand-Receptor interactions. This work was led by @artofbiology (Min Dai in my lab) with Fei Chen's lab at the Broad. www.biorxiv.org/content/10.1...

And now the Neuromorphic piece has been picked up a one of the 10 most notable AI papers this week! Exciting! Well done Suraj!!
aiworld.eu/story/this-w...

Well to complement my son's efforts here is some background into my own first foray into neuromorphic computing. The original article can be seen in PNAS! A big shout out to Suraj Honnuraiah, who made it all happen! www.growkudos.com/publications...

SNNs Are Not Transformers (Yet): The Architectural Problems for SNNs in Modeling Long-Range Dependencies Spiking Neural Networks (SNNs) have gained increasing attention for their ability to operate on low-power neuromorphic hardware, offering a pathway toward energy-efficient large language models (LLMs)...

Well when it rains it pours. My son Will @fishell_will , just published a paper on biorxiv, much closer to my heart with my lab member Suraj Honnuraiah on spiking neural networks and understanding why they struggle as LLMs. www.biorxiv.org/content/10.1...

In a whimsical review of his arXiv.org paper, my son Will discusses the use of TempoBench in evaluating LLMs. I particularly enjoyed the faux ad for the TempoBench "product". You go Will! wfishell.github.io/landing.html

Mechanics of Learned Reasoning 1: TempoBench, A Benchmark for Interpretable Deconstruction of Reasoning System Performance Large Language Models (LLMs) are increasingly excelling and outpacing human performance on many tasks. However, to improve LLM reasoning, researchers either rely on ad-hoc generated datasets or formal...

Really excited to see my son Will @fishell_will publish his first work Tempobench that measures temporal causality and reasoning in LLMs. arxiv.org/abs/2510.27544

New updated on how pyramidal neurons impact interneuron subtypes. Turns out spontaneous activity, Wnt signaling in PV Interneurons and integrin-signaling in SST neurons are the culprits! Also see how optimal transport tracks identity.https://www.biorxiv.org/content/10.1101/2024.07.20.604399v2

I am so excited about this combined analysis of the selective CRHR2 population within the lateral septum and how they control defensive behavior. Dionnet was truly a gifted graduate student. Keep your eyes on this one!

At my last year of attending the McKnight conference. Such an amazing group of scientists. A big shout out to Lauren Orefice who gave an amazing talk this morning!

Here is a chat with Bosilijka Tasic and myself about the enhancer-mediated AAVs that we were both centrally involved in publishing. I truly believe these approaches are fundamentally important for both basic and clinical approaches! braininitiative.nih.gov/news-events/...

The U.S. Lit a Beacon for Science. Under Trump, Scientists Fear It’s Dimming

So I open up the NYtimes and there is my pal @ardemp.bskyverified.social saying that despite funding cuts he is staying in this country that he loves. Keep the faith you all!.... www.nytimes.com/2025/06/03/u...

Wow quite the legends. Not to mention the white asparagus. Enjoy your fragrant pee dude!

Exploring brain circuits, one cell type—or more— at a time This issue of Neuron, with accompanying papers in Cell, Cell Genomics, Cell Reports, and Cell Reports Methods, contains studies describing methods for access to neuronal and non-neuronal cell types wi...

"Exploring brain circuits, one cell type—or more— at a time" www.cell.com/neuron/fullt...

Bosiljka Tasic & @gordfishell.bsky.social ‬
@cp-neuron.bsky.social

Yep, glad people in europe are taking notice. Hey guys, we are having a giant fire sale! Top scientists looking for relocation to the other side of the pond. Sad!

One thousand candidate enhancers tested in vivo in the mouse brain! A massive resource and oh so useful as validation set for genome-wide enhancer prediction methods. Super fun to be involved in one of the papers: ‘the prediction challenge paper’ by Nelson&Niklas et al www.cell.com/cell-genomic...

Here is the collective effort of 12 labs, lead by my own at the Broad to target the range of cortical interneuron subtypes! These tools should be a game changer for all of you using NHP and other less genetically amenable species! Have at it!
urldefense.proofpoint.com/v2/url?u=htt...

Check out the neuroview and all the great articles of new tools to target different cell types with enhancer-mediated AAVs!

Even though it is spring....it seems to have gotten dark lately...."Within the last several days, Harvard University has received letters from most federal agencies indicating that nearly all of HMS’s active, direct federal grants have been terminated."

There you are with my buddy Domingos!!!

You should write Kara directly or maybe just watch for it on Bluesky. For me this is a work in progress but I will post on Bluesky as I get involved.

I will fly to anywhere in the United States on my own dime to talk to people about basic science! I encourage all of you, particularly senior scientists to make this commitment. If you are looking for material ask @karalmarshall.bsky.social She has put together an amazing basic talk!

OK inspired by @karalmarshall.bsky.social I am reaching out to the public where ever I can. Traditionally Republicans have been incredibly supportive of science. We as scientists need to reach out to people across the spectrum to rally their support for basic science!

Ask Kara for this talk! It is an amazing walk through for the general public.mmI am going to give a general talk to the public every time I am traveling through the USA. I have had a wonderful career. It is my students and postdocs I am worried about 🥺. Come help out!

Yes I know....I was hoping you guys would overlook the error!