Why ChatGPT Suddenly Has a Thing for Goblins

OpenAI has addressed a curious trend: ChatGPT users have noticed the AI model frequently referencing goblins in its responses. The phenomenon, documented by Mashable, appears to stem from the model's training data, which includes a disproportionate number of goblin-related examples. OpenAI explained that the model's outputs reflect patterns in the data it was trained on, and the goblin references are not intentional but rather a statistical artifact. The company is monitoring the situation and may adjust training data to reduce such quirks in future updates.