Language-Models | Sebastian Spicker

2026

1 article

4 Mar 2026 The Model Has No Seahorse: Vocabulary Gaps and What They Reveal About LLMs There is no seahorse emoji in Unicode. A language model may nevertheless claim there is one or emit a plausible substitute. That failure is real, but the output alone does not tell us whether tokenisation, training data, prompting, decoding, or verification caused it. ailanguage-modelstokenisation

2025

1 article

20 Mar 2025 The Oracle Problem: What The Matrix Got Right About AI Alignment The Oracle is the most interesting character in The Matrix for anyone who thinks about AI alignment. On one reading, she manages Neo’s beliefs for an outcome she judges good. That makes her a useful fictional case about honesty, autonomy, and paternalism—not a textbook diagnosis of a real AI system. aialignmentfilm

2024

1 article

7 Oct 2024 Three Rs in Strawberry: What the Viral Counting Test Actually Reveals The viral “three Rs in strawberry” test exposes a real mismatch between token-level input and character-level operations. It does not, by itself, identify one tokenizer boundary or explain any particular model response. llmtokenisationlanguage-models