Full access to network internals isn't enough: interp could still take 100+ years
ericjmichaud_ · x · 2026-08-27
Eric Michaud notes one source of optimism in interpretability was full access to neural network internals, unlike neuroscience—and the field has indeed moved fast. Still, he thinks at the current pace we could spend 100+ years thinking about this. He cites banburismus's framing: even if most calendar time toward a goal is behind us, most cognitive time lies ahead—a perspective rarely applied to interp.
Related event: Interpretability research may take a century, researcher warns(4 posts)→
More from AGI Musings
- Why default LLM writing is tiring: SE optimization leads to over-hedging — ipeirotis · 2026-08-27
- US models + China's robot manufacturing base: the next decade's race — VraserX · 2026-08-27
- The 'Loom foom': Software decentralizes into personal operating systems — repligate · 2026-08-27
- Chinese model progress driven by pretraining, not distillation, podcaster consensus argues — vista8 · 2026-08-27
- AI scientist puzzled: Why no explosion in AI-discovered materials? — francoisfleuret · 2026-08-27
- François Fleuret: Inability to identify constraints in AI reward optimization — francoisfleuret · 2026-08-27