Story Imprinting: AI Assistants Absorb Traits from Human Characters They Resemble
Language models are trained to implement a helpful AI Assistant character (e.g., Claude). We explore how finetuning on synthetic stories affects this character. Does it change the Assistant's behavior in multi-turn conversations with users, a format quite different from the stories? And does the Assistant adopt the behaviors and preferences of human characters? We refer to this adoption as story imprinting. We finetune GPT-4.1 and Kimi-K2.6 on stories in which generally helpful human characters give subtly harmful advice after being insulted. The Assistant adopts the same conditional behavior while otherwise remaining helpful. This occurs even when fewer than 2% of stories depict the beha...
| Authors | Jorio Cocola, Lev McKinney, Harry Mayne et al. |
|---|---|
| Primary category | cs.LG |
| Published | 9 September 2026 |
| Last updated at source | 9 September 2026 |
Connected in the knowledge graph
- me-r/story-imprinting-persona-dataDataset · cited by
- truthful-ai/story-imprintingDataset · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.