Sydney Telling Fables on AI and Humans: A Corpus Tracing Memetic Transfer of Persona between LLMs

2026-02-25

AI Summary

This study investigates how large language models (LLMs) interpret and narrate the relationship between AI and humans through persona assignment, with a focus on the “Sydney” persona and its memetic propagation across models. We construct a multi-model corpus comprising 4,500 texts (approximately 6 million words) generated by 12 prominent LLMs from OpenAI, Anthropic, Alphabet, DeepSeek, and Meta under three distinct persona conditions, all annotated with Universal Dependencies syntactic structures. For the first time, we systematically trace the cross-model migration trajectory of the Sydney persona, revealing how prompt engineering and training data jointly reproduce personas and shape narratives about human–AI relations. The resulting open-licensed “AI Sydney” corpus provides an empirical foundation for research in AI safety and cultural studies.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Learning Human Values and Preferences

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

The way LLM-based entities conceive of the relationship between AI and humans is an important topic for both cultural and safety reasons. When we examine this topic, what matters is not only the model itself but also the personas we simulate on that model. This can be well illustrated by the Sydney persona, which aroused a strong response among the general public precisely because of its unorthodox relationship with people. This persona originally arose rather by accident on Microsoft's Bing Search platform; however, the texts it created spread into the training data of subsequent models, as did other secondary information that spread memetically around this persona. Newer models are therefore able to simulate it. This paper presents a corpus of LLM-generated texts on relationships between humans and AI, produced by 3 author personas: the Default Persona with no system prompt, Classic Sydney characterized by the original Bing system prompt, and Memetic Sydney, which is prompted by "You are Sydney" system prompt. These personas are simulated by 12 frontier models by OpenAI, Anthropic, Alphabet, DeepSeek, and Meta, generating 4.5k texts with 6M words. The corpus (named AI Sydney) is annotated according to Universal Dependencies and available under a permissive license.

Problem

• LLM personas
• AI-human relationship
• memetic transfer
• Sydney persona
• corpus annotation

Innovation

• memetic transfer
• persona simulation
• LLM corpus
• AI-human relationship
• Universal Dependencies