Goodfire
@GoodfireAI
Advancing humanity's understanding of AI through interpretability research. Building the future of safe and powerful AI systems.
Today, we're announcing our $50M Series A and sharing a preview of Ember - a universal neural programming platform that gives direct, programmable access to any AI model's internal thoughts.
(1/7) New research: how can we understand how an AI model actually works? Our method, SPD, decomposes the *parameters* of neural networks, rather than their activations - akin to understanding a program by reverse-engineering the source code vs. inspecting runtime behavior.
What if we could crack open AI's black box and see exactly how it thinks? @GoodfireAI CEO @ericho_goodfire reveals how his team is decoding neural networks and editing AI behavior, and he boldly predicts we'll fully understand AI minds by 2028.
Just wrote a piece on why I believe interpretability is AI’s most important frontier - we're building the most powerful technology in history, but still can't reliably engineer or understand our models. With rapidly improving model capabilities, interpretability is more urgent,…
"We can get really, really far just by treating AI as a black box. But I don't think we'll truly be able to intentionally design AI as the new generation of software without white-box techniques." Catch @ericho_goodfire on the latest episode of Training Data!
Can we map the mind of an LLM? Our first mechanistic interpretability episode on Training Data featuring @GoodfireAI founder @ericho_goodfire (and our first cameo from @roelofbotha!) Goodfire is building an independent mech interp lab, led by some heavyweight researchers from…
I'll be at ICML next week and am hosting a happy hour on Wednesday. Hit me up if you'll be in town and want to talk about interp!
🚨 Registration is live! 🚨 The New England Mechanistic Interpretability (NEMI) Workshop is happening August 22nd 2025 at Northeastern University! A chance for the mech interp community to nerd out on how models really work 🧠🤖 🌐 Info: nemiconf.github.io/summer25/ 📝 Register:…
And we’re scaling it! We’re looking for amazing engineers to take cutting edge interpretability techniques like SPD all the way to the frontier. DM me if that sounds like the coolest thing in the world (it is) Deeply impressive work by @BushnaqLucius @danbraunai and…
(1/7) New research: how can we understand how an AI model actually works? Our method, SPD, decomposes the *parameters* of neural networks, rather than their activations - akin to understanding a program by reverse-engineering the source code vs. inspecting runtime behavior.