←back to thread

AI models need a virtual machine

(blog.sigplan.org)
224 points azhenley | 2 comments | 30 Aug 25 13:25 UTC | HN request time: 0.411s | source
Show context
mehulashah ◴[30 Aug 25 14:30 UTC] No.45074995[source]▶
I do believe it’s time for systems folks to take a hard look at building systems abstractions on top of LLMs as they did 50 years ago on top of CPUs. LLMs are the new universal computing machines, but we build with them like we built computers in the 1950s - one computer at a time.
replies(6): >>45075043 #>>45075096 #>>45075104 #>>45075134 #>>45075169 #>>45075192 #
1. moduspol ◴[30 Aug 25 14:55 UTC] No.45075192[source]▶
CPUs are deterministic! I think it’s going to be inherently limiting how far you can abstract above something non-deterministic.
replies(1): >>45077015 #
2. mehulashah ◴[30 Aug 25 18:45 UTC] No.45077015[source]▶
Agree. The non-determinism is a problem. Perhaps systems thinking can help here. After all, traditional computer systems often behave non-deterministically and we’ve learned to make them more reliable.