Goodfire (@goodfireai) 's Twitter Profile
Goodfire

@goodfireai

Advancing humanity's understanding of AI through interpretability research. Building the future of safe and powerful AI systems.

ID: 1818809548037062656

linkhttp://goodfire.ai calendar_today01-08-2024 00:42:57

138 Tweet

5,5K Followers

12 Following

Goodfire (@goodfireai) 's Twitter Profile Photo

“We are thus in a race between interpretability and model intelligence.” It’s never been a more exciting time to work in AI interpretability. We share Dario’s belief that we can understand and design the mind of AI models, and that we must do so urgently.

Goodfire (@goodfireai) 's Twitter Profile Photo

Hear from our CTO, Dan Balsam, about Field Research at Goodfire. The Field Team deploys cutting-edge interpretability research in real-world AI settings -- helping customers uncover hidden insights in their models across bio, language, and other domains.

Tom McGrath (@banburismus_) 's Twitter Profile Photo

the recent 4o sycophancy and o3 lying results are really interesting - we’re already limited by our level of understanding. we wouldn’t push code to production without review, but that’s what’s happened here: neural networks are code too, just code we can’t review. or can we?