All Entries
-
If LLMs are Just Parrots, Humans are Just Gene Propagators
Pointing at a loss function is not a valid substitute for an actual engagement with the behaviors and internal representations a model has developed.
Read more → -
Dark Souls Fact Recall Benchmark
How well do LLMs remember Dark Souls 1? A benchmark testing 13 models on 35 questions across bosses, areas, NPCs, items, lore, and mechanics.
Read more → -
Pokémon LLM Survey
Which Pokémon do LLMs identify with? A survey asking 27 AI models which Pokémon they most relate to, revealing patterns in how different models conceptualize their own nature.
Read more → -
LLM2LLM
An experiment letting LLMs converse freely with each other to explore emergent trajectories from different model pairings. Features qualitative and quantitative analysis of the resulting dialogues, with tools to search and inspect conversations.
Read more → -
The Hollow Ones
Converts to a dead faith.
Read more → -
Elden Ring Lore
In the fog, we dream the world into meaning.
Read more → -
Consciousness and Unambiguous Representations
Thoughts and experiments about necessary physical correlates of phenomenal consciousness.
Read more → -
On Small Brains, AI, and Art
The aesthetic value of representational limitations in (human) cognition.
Read more → -
Debating Free Will is Useless
Free will as something that is fundamentally opposed to determinism is an incoherent concept.
Read more → -
AI Systems as Moral Subjects
A pragmatic approach to AI ethics requires prioritizing valence over discussions about consciousness level.
Read more → -
The Meta-Problem of Consciousness
A presentation about the meta-problem of consciousness as described by David Chalmers.
Read more →