---
id: 1653
title: "Yes. LLMs can create convincingly human output."
source_domain: josephthacker.com
source_url: "https://josephthacker.com/ai/2023/08/30/humanlike-llm-ouput.html"
published_at: "2023-08-30T00:00:00Z"
tags: ["ai-security", "llm", "prompt-engineering", "prompting", "rlhf", "ai-writing", "tone"]
phying_url: "https://phying.com/item/1653"
lang: en
---

# Yes. LLMs can create convincingly human output.

## Summary

Joseph Thacker explains why RLHF makes LLMs default to a formal academic tone, and shares prompting strategies—informal requests, vocabulary limits, style examples—for human-like output.

## Intelligence Metadata

- **Sources:** josephthacker.com
- **Published:** 2023-08-30T00:00:00Z
- **Original:** https://josephthacker.com/ai/2023/08/30/humanlike-llm-ouput.html
- **Phying News:** https://phying.com/item/1653