In the late 19th century, some believed that physical features could indicate a person's character, a notion supported by criminologist Cesare Lombroso. Today, readers identify AI-generated writing by looking for specific stylistic elements, such as the use of em dashes. However, these features do not definitively indicate the author's identity. As AI models improve, their ability to mimic human writing will also enhance. Some users have started to make intentional spelling errors to distinguish themselves from AI.
Every writer has a unique style that reflects their personal experiences. For example, Jane Austen and Charles Dickens both excelled in comedy, yet their styles differ significantly due to their backgrounds. AI writing, on the other hand, lacks a fixed style, as it generates text based on learned patterns from extensive training on various texts. This probabilistic approach affects how AI tells stories.
Experiments with ChatGPT reveal its tendency to produce fables or moralistic tales, often incorporating clear lessons. For instance, when prompted to tell a story about the triumph of evil, ChatGPT created a narrative about a city built around a clock, emphasizing themes of control and freedom. Despite attempts to guide the AI towards less didactic storytelling, it often defaults to familiar tropes and moral conclusions.
AI-generated writing reflects societal expectations, as it reproduces patterns from existing texts. While this can be comforting, it also limits creativity. Genuine innovation in writing stems from personal experiences and unique perspectives, which AI lacks. The rise of AI writing may challenge traditional literary forms but could also inspire human writers to explore new creative avenues, similar to how photography influenced painting in the 19th century.
Ultimately, AI writing serves as a reminder of the value of authentic human expression, encouraging writers to convey their experiences and insights in innovative ways.