Goodfire

‘We Were Looking at It Wrong’: Goodfire Researcher’s Claim Sends AI Industry Into Reassessment

A sweeping blog post by Goodfire researcher Nick Cammarata has ignited intense debate across the AI sector after arguing that the field may have fundamentally misunderstood how neural networks should be interpreted and studied. The claim, framed as “good news about the nature of neural networks,” has triggered a wave of speculation because it suggests the internal structure of modern AI systems may be more legible, and more controllable, than many researchers assumed.