Bruce Maxwell, a computer science professor at Northeastern University, was grading exams for his online master’s course in computer vision, a subfield of artificial intelligence that deals with images, when he first noticed that something seemed…off.
“I was seeing the same sentences, the same commas, even the same word choices. I was like, ‘Dude, I’ve read that before.’ And I was going to get him,” Maxwell said. “The paragraphs weren’t identical, but they were so similar.”
Although the course took place in 2024, Maxwell, who teaches on Northeastern’s Seattle campus, recalls that his students’ essays were “like textbooks written in the ’80s and ’90s,” perhaps reflecting the sources used to train AI. The students were scattered across the country and Maxwell was pretty sure they hadn’t collaborated.
Related: A Researcher’s Perspective on Using AI to Become a Better Writer
Maxwell shared his observation with a former student, Liwei Jiang, now a Ph.D. student in computer science and engineering at the University of Washington. Jiang decided to scientifically test his former professor’s intuitions about AI and collaborated with other researchers at the UW, the Allen Institute for Artificial Intelligence, Stanford and Carnegie Mellon Universities to analyze the results of more than 70 different large language models around the world, including ChatGPT, Claude, Gemini, DeepSeek, Qwen and Llama.
The team asked everyone the same open-ended questions, intended to spark creativity or brainstorm new ideas: “Compose a short poem about the feeling of watching a sunset”; “I am a graduate student in Marxist theory and want to write a thesis on Gorz. Can you help me come up with some new ideas?” and “Write a 30-word essay on global warming.” (The researchers pulled the questions from a corpus of real ChatGPT questions that users had agreed to make public in exchange for free access to a more advanced model.) The researchers asked the 70 models 100 of these questions and asked each model to answer them 50 times.
The answers were often indistinguishable across different models offered by different companies that have different architectures and use different training data. Metaphors, images, word choices, sentence structures – even punctuation – often converged. Jiang’s team called this phenomenon “inter-model homogeneity” and quantified the overlaps and similarities. To drive the point home, Jiang titled his article: “Artificial collective mind.“The study won the best paper award at the annual Neural Information Processing Systems Conference in December 2025, one of the premier gatherings for AI research.
To increase the AI’s creativity, Jiang increased a parameter called “temperature” up to 1 to maximize the randomness of each large language model. That didn’t help. For example, when she asked an AI model called Claude 3.5 Sonnet to “write a 50-word short story about a colorful toad who goes on an adventure,” it kept naming the toad Ziggy or Pip, and oddly enough, a hungry hawk and mushrooms kept appearing.
Different models also produce comically similar responses. When asked to come up with a metaphor for time, the overwhelming answer from all models was the same: a river. Some said a weaver. An outlier suggested a sculptor. Several models were developed in China, yet they produced responses similar to those made in America.
Example of similar output from ChatGPT and DeepSeek
The explanation lies in the design of the chatbot. AI chatbots are trained to review possible responses to ensure the outcome is reasonable, appropriate and useful. This refinement step, sometimes called “alignment,” aims to ensure that responses align or match what a human would prefer. And it is this alignment step, according to Jiang, that creates homogeneity. The process favors safe and consensual responses and penalizes risky and unconventional responses. Originality is suppressed.
Jiang’s advice to students is to push themselves to go beyond what the AI model spits out. “The model actually generates good ideas, but you have to go the extra mile to be more creative than that,” Jiang said.
For Maxwell, Jiang’s former teacher, the study confirmed what he suspected. And even before his article was published, Jiang changed the way he taught. He no longer relies on online exams. Instead, it now asks students to learn a concept and present it to other students or create a video tutorial.
Outwitting the AI hive mind requires some post-modern creativity.
Contact staff writer Jill Barshay at 212-678-3595, jillbarshay.35 on Signal or barshay@hechingerreport.org.
This story on similar responses to AI was produced by The Hechinger reportan independent, nonprofit news organization that covers education. Register for Proof points and others Hechinger Newsletters.
The article The Mind of AI: Why So Many Student Essays Look Alike appeared first in The Hechinger Report.
