Interpolation and hallucination: why sampling outside training data yields AI slop
gerardsans · x · 2026-10-23
The author frames hallucinations as navigating sparse or out-of-distribution regions of latent space via embedding interpolation — the familiar ghostly images and wrong-fingered hands. He argues frontier AI is fundamentally sampling-based: outputs derive from training data samples, so new combinations of existing data are possible, but nothing genuinely new can be created outside that distribution by AI alone.
Related event: Interpolation in Sparse Latent Space May Explain AI Hallucinations(2 posts)→
More from Models
- Ethan Mollick: AI task horizon went from 4 minutes to 17 weeks between his two books — emollick · 2026-10-23
- OpenAI's Dot UX: Speech Is Linear, Real Value Is Multithreaded — cocktailpeanut · 2026-10-23
- Sam Altman: models now find bugs outside any hacker threat model, 'we can't just patch them all' — ns123abc · 2026-10-23
- Omnii Genome: single generalist model beats AlphaGenome at linking variants to gene activity — BenSiranosian · 2026-10-23
- MemTensor ships MemT-1.0 nano and flash with a new parametric memory design — Alarming-Ad8154 · 2026-10-23
- Axion Labs launches non-transformer decision models for efficient edge reasoning — tw_killian · 2026-10-23