Test finds base model completions inherit AI/human writing traits from prompts
AaronBergman18 · x · 2026-08-20
An interesting test on Pangram and base models revealed that if the prompt text is AI-generated, the completion text is reliably flagged as such; conversely, if the prompt is human-written, the completion is flagged as human-written. This suggests the model inherits stylistic traits from the prompt during completion.
More from Models
- Sanity Check: Claude $100 Seat Equals ~$2500 in API Credits for Heavy Users — weed_cutter · 2026-08-21
- Claude accurately predicted Qwen 3.8 27B performance benchmarks — OneMoreName1 · 2026-08-21
- GLM 5.3 Scores 47.1% on SlopCodeBench, Ties with Fable 5 — corruptbytes · 2026-08-21
- Post-training causes LLMs to produce novel but impractical language — TuhinChakr · 2026-08-20
- User says Grok 4.6 now handles 100% of coding work previously done with Codex — CedricMakes · 2026-08-20
- Grok 4.6 leads in legal/GDP benchmarks, lags in coding — ChrisGPT · 2026-08-20