Ethical Manifesto Released for AI-Assisted Writing

A group of professionals has introduced the ‘Ethical Manifesto of Writing with Artificial Intelligence,’ a ten-principle framework designed to guide authors using AI tools. Proposed by Fulvio Julita and Alessandra Perotti, the manifesto was developed by a committee of eleven experts spanning journalism, law, publishing, and management.

The core philosophy emphasizes that the human author must remain responsible for the final output. Key principles include the necessity of human fact-checking to combat AI hallucinations—instances where models generate plausible but false information—and the requirement that authors maintain control over tone and privacy. The manifesto asserts that even the most elaborate prompts cannot replace accurate research or the human responsibility of a signature.

Technical research from OpenAI highlights that these errors occur because language models are designed to predict the most statistically probable next word rather than verify truth. Consequently, the manifesto urges writers to use AI as a tool for efficiency while ensuring that the final text remains a product of human oversight and accountability.