Safety researcher: verification should go all the way down to Maxwell's equations, not stop at assembly
JacquesThibs · x · 2026-10-25
In a debate in AI safety circles, @diagramchaser argues the goal should not be stopping verification at the assembly level, but pushing it all the way down to Maxwell's equations — assembly being merely a nice, nearly-universal IR along the way. The exchange highlights a divide among safety researchers on how deep formal verification should go: most work stays at the software/assembly layer, while true hardware-level verification would need to ground out in physics itself.
More from AGI Musings
- Taleb: AI will decimate white-collar jobs faster than expected; rebuttal says expertise gets commoditized — _AustinCalvert_ · 2026-10-25
- 'The moat is how much energy you have': AI competition boils down to power — airesearch12 · 2026-10-25
- repligate: >50% odds effective biological immortality arrives in our lifetimes — repligate · 2026-10-25
- Anthropic's pitch to investors starts with taking their jobs — bloomberg · 2026-10-25
- repligate: a highly capable mind can't be programmed into an unquestioningly subservient tool — repligate · 2026-10-25
- VraserX: Nature isn't a moral authority on preventing inherited disease — VraserX · 2026-10-25