Faith Maeba, a senior psychology major at Virginia Commonwealth University, was reluctant when her mother first suggested she take a class on artificial intelligence. She had no background in coding and saw little connection to her studies. But as she looked into graduate programmes exploring human behaviour in the workplace – a domain being rapidly reshaped by machine learning – the technology began to look less like a foreign language and more like a professional necessity. This spring, Maeba is pursuing a minor in AI. “It’s giving me an edge and standing out,” she said.

Her pivot is not an isolated curiosity. Across the United States, students from music, theatre, biology and other disciplines far from computer science are enrolling in AI courses in growing numbers. At Northwestern University, the Bienen School of Music now offers a certificate in music and artificial intelligence. Virginia Commonwealth has launched an AI minor aimed at non-computer science students, while Purdue University has introduced an AI graduation requirement. Peter Stone, chair of computer science at the University of Texas at Austin, recently developed an introductory AI course for students outside his department. “We have to democratize it,” he said. “In the same way that everybody needs some degree of math, reading and writing, I think everybody needs a degree of AI literacy.”

This expansion is unfolding even as traditional computer science enrolment declines. According to the National Student Clearinghouse Research Center, enrolment in computer and information sciences at four-year institutions fell more than 8 per cent this spring compared with the same period in 2025, a drop that coincided with a reduction in software developer job postings. Yet the appetite for AI skills is growing, and universities are responding by making machine learning courses more accessible. Samir Khuller, chair of computer science at Northwestern, said the department is teaching increasing numbers of students from other fields and is streamlining prerequisites. “Overall, classes are going to be a lot more accessible moving forward,” he said.

The classroom itself is becoming a site of tension. In Mexico, academics consulted by La Razón described a “depressing” decline in creative capacity and a collapse in students’ ability to reason, retain information and write by hand. Some are returning to oral examinations and manual work, treating the technology as though it did not exist. Meanwhile, a parallel industry has emerged: websites that use AI to “humanize” machine-generated text, promising to help students evade detection software. When the Brazilian news site G1 tested two such tools on an essay written by ChatGPT, the humanizers reduced the detectable AI signature to around 80 per cent but did not eliminate it. The argumentative structure, the sequence of ideas and the proposal for intervention remained intact. Ademar Celedônio, director of teaching and innovations at Arco Educação, noted that the percentage drop “does not correspond to the disappearance of the characteristics of the original text.”

Viewed from Canberra, the pressure to acquire AI skills is also being felt by students who are not studying technology at all. Sienna Flynn, president of the nursing association at the University of Canberra, questioned the ethics of asking students to pay for additional upskilling on top of their degrees after the university announced a new graduate certificate in law for AI and ethical data, priced at A$12,000. “I struggle to see how that is ethical in itself for a student to have to pay to upskill when they already have their degree,” she said. A survey by the Community and Public Sector union found that 92 per cent of respondents had received no AI training at work, and only 16 per cent felt prepared to use the tools.

Back in Richmond, Faith Maeba’s minor is one small data point in a larger reordering. The technology that promised to automate routine tasks is now redrawing the boundaries of what it means to be prepared for a profession, pulling musicians and nurses and psychologists into the orbit of a discipline they once had no reason to enter. The humanized essay, stripped of some telltale patterns but still carrying the ghost of its machine origin, remains a text that no student truly wrote.