Very intersting read: Nick Cammarata, a co-author of the 2020 Zoom In paper that introduced the idea that neural networks fold unrelated concepts into the same neurons, now says the field misread its own key evidence. In an essay published Thursday by interpretability startup Goodfire, he reports that in a preliminary audit of Qwen3-32B, roughly 90% of neuron signals have one main job and only 2-4% are not yet understood; the automated study he started at OpenAI in 2023 judged about 1.7% legible with the tools of the time. After all, let’s remember: According to OpenAI’s own System Card, GPT-6 Astra's reasoning is harder to monitor than that of its predecessors. According to The Information, the next model is expected to rely on Loop Transformers, which think more in numbers and less in readable text. That sounds like a solution: if the text disappears, you just look inside. However, this is still far from being a viable real-time monitoring method, and the testing was conducted on classic transformers, not on loop architectures. Still, it's a good step in the right direction and can be a great help in understanding the system, and thus in ensuring its safety.

Tommy Shaughnessy says an unnamed Anthropic researcher told him over dinner: “I’m just enjoying my last six months.” Speaking with Michael Arrington, he said the researcher was serious and claimed most Anthropic employees share that outlook. The problem, of course, is that it's unclear what this means. I'm guessing they mean that *scientific work* will be completely taken over by feedback loops, and we'll disappear into RSI. I'm not reading this as a doomsday post, but rather as a positive outlook that AI supports and transforms all areas.