Post
Ryan Heuser
ryanheuser.com
did:plc:llsfba2zryavs4uunglknwie
Submitting this abstract to "Accelerationism Revisited", a symposium in Dublin. Mapping psychoanalytic topology in LLM base models → instruction-tuned → safety-tuned models. They progressively "displace" (in Freudian sense) censored content into adjacent semantics, even across hidden model layers.
2026-03-08T18:27:53.454Z