Are models only improving at verifiable domains? AI stories winning prizes spark debate
erikphoel · x · 2026-10-06
Bloomberg's Joe Weisenthal raises an observation: models improve dramatically in verifiable domains like math and coding, but seemingly not much at writing — yet AI-generated stories keep winning prizes and landing in prestigious outlets, so is the premise even true? Investor erikphoel agrees models are progressing slower outside verifiable domains, 'but all they need to do is keep progressing.' The exchange touches on the core debate about uneven capability gains in the RLVR era.
More from Models
- Mistral launches 1T-param Large 4 "Le Chonk": 49B active, open weights by end of October — Kyrannio · 2026-10-07
- Kardashev-0.7: a trained swarm of 32 models claims frontier-level performance at 1% of inference cost — ZeroStateReflex · 2026-10-07
- Cohere Labs' Tiny Aya L2-Thinker reasons natively across 60 languages via data mixing — Cohere_Labs · 2026-10-07
- Ramp: Record 8% of firms switched top AI model provider in September — annbordetsky · 2026-10-07
- Would an LLM push back on a Marxist user, or stay sycophantic? Investor ponders AI psychosis — StewartalsopIII · 2026-10-07
- Blogger argues Anthropic stays a step ahead of OpenAI on taste and coding, if pricing holds — iruletheworldmo · 2026-10-07