Safety
this is the whole problem in a nutshell. LLM-centered systems just can’t be trusted to follow hard constraints, we absolutely must find alte…
this is the whole problem in a nutshell. LLM-centered systems just can’t be trusted to follow hard constraints, we absolutely must find alternatives that can, or we are screwed. Anybody remember this
this is the whole problem in a nutshell. LLM-centered systems just can’t be trusted to follow hard constraints, we absolutely must find alternatives that can, or we are screwed. Anybody remember this whole marketing spiel about Constitutional AI and how it's supposedly the safer path to AGI? ... Screenshot from the current website. ... That approach is clearly not working
Source: Gary Marcus (X) | 2026-07-31