When university education is fully integrated into AI systems, what will happen next? In February 2025, California State University (CSU) and OpenAI reached a $16.9 million cooperation agreement. OpenAI opened ChatGPT Edu to more than 500,000 students and faculty within the university system, which is the largest deployment of ChatGPT application by a single institution in the world.
This started the largest AI experiment in the history of American universities, and AI began to play a series of new roles on campus: AI administrators appeared in the library, the career center used AI to guide students in their job search, ChatGPT was integrated into the teaching, classroom and graduation ceremony processes, and the digital avatar also replaced the principal in delivering speeches and communicating at the orientation ceremony...
But the plan has also aroused opposition from many faculty and staff, who worry that AI will replace humans sooner or later, further exacerbating the education crisis. Students are also full of doubts about this plan. They don’t know what role AI can play in their learning and growth. Nonetheless, CSU recently signed a new three-year agreement with OpenAI worth $13 million per year.
A question worth pondering is: Is AI a powerful tool to help students develop better, or does it cause them to lose their critical thinking skills?
American university launches largest AI experiment
The artificial intelligence program launched by CSU covers its 23 campuses, 460,000 students and 63,000 faculty and staff. As the nation’s largest four-year public higher education system, it hopes the initiative will help it become the nation’s largest artificial intelligence-driven public university system.
As the oldest campus among the CSUs, San Jose State University not only pioneered the establishment of a full-time AI librarian, but also established an AI education and innovation collaborative space for the public, faculty and students - the Center for Artificial Intelligence for Citizenship and Social Good.
In the fall of 2025, the school released the "AI Everywhere" strategy and made AI literacy training a required course for freshmen. The school’s business school runs an AI boot camp for high school students, and Adobe sponsors the AI Career Development Center. In addition, the AI assistant helped coordinate various aspects of the graduation ceremony. What’s even more interesting is that the principal used an AI digital clone to deliver a speech and communicate with new students, parents and alumni at the orientation ceremony.
The development of AI technology has the potential to open up a new situation in the future job market. According to a recent report, about 40% of recent college graduates in the United States are underemployed. Most of the students in the CSU system are first-generation immigrants or the first in their families to go to college. In CSU’s view, AI will be a key driver for working-class students to keep pace with the development of the AI economy.
California has natural conditions as a test field for artificial intelligence education. Many of the world's largest technology companies have their headquarters in the state. The California government has also provided support and assistance for the development of AI. In August 2025, California Governor Gavin Newsom announced an education cooperation agreement with Adobe, Google, IBM, and Microsoft. Each company will provide free artificial intelligence resources to California universities, aiming to gradually build the future artificial intelligence talent team by training high school students, community college students, and CSU students to use AI.
Laying off employees while buying AI: Why did universities erupt in opposition?
According to the cooperation agreement, CSU’s previous cooperation agreement with OpenAI will expire on June 30 this year, and this new signing is a new three-year agreement worth $13 million per year. Despite the good intentions of introducing AI technology as part of the campus system, the plan surprisingly sparked strong opposition from CSU faculty and staff.
The petition signed by faculty states that this is partly because faculty believe that, aside from privacy and security features, ChatGPT Edu is not fundamentally different from the free online version of ChatGPT. In addition, the full use of AI systems also involves issues of academic integrity and fair use. "Although the system is called ChatGPT Edu, it is a general-purpose chatbot and is not designed, trained, or optimized for education."
In fact, faculty and staff are facing a dilemma: On the one hand, CSU is facing a $2.3 billion funding gap and a large number of faculty are facing layoffs. In the past two years, San Francisco State University has eliminated 615 lecturer positions. Some schools have had entire academic departments closed and tuition fees increased by 6%.
(Source: CSU Campus)
The issue of funding priorities was one of the key factors in faculty opposition to contracting with OpenAI. They believed that the budget should be used to support faculty, but apparently, CSU chose to go contrary to their expectations and prioritize the budget for AI technology.
"We believe investing in CSU's human resources is the best way to ensure the quality of research, teaching, and learning in California's public education," they wrote in the petition. "The challenges we face in higher education cannot be solved by artificial intelligence. Instead, we must empower faculty, staff, and students to work together to build a sustainable, people-centered future for CSU."
Currently, CSU has activated about half of its ChatGPT licenses, which means the school has paid for at least tens of thousands of accounts. On the one hand, there is a fiscal deficit, and on the other hand, there is heavy investment in AI technology, so some people think this behavior is "absurd."
On the other hand, the plan was not consulted with faculty in advance. Although the school does not force faculty to use it, it clearly states that "refusing to integrate artificial intelligence into the curriculum is like swimming against the tide." Therefore, faculty use AI more passively. They are afraid of losing their jobs and do not know what AI can do in teaching and research.
The more important question: What is left of education?
The backlash got CSU’s attention. A recent survey of more than 94,000 students and staff at the school showed that 52% of professors believe AI has had a negative impact on their teaching, and 67% of students believe they are not being taught how to use AI effectively.
Policies regarding the use of the system are unclear. CSU provides tools and training on a website called AI Commons, but leaves it up to each school to decide how to use the chatbot. But these resources appear to have not been implemented, with data showing that as of April this year, only 0.7% of students and 16% of staff had completed voluntary training.
A practical question is, if students use AI in submitting assignments, how to define the student’s contribution? In teaching, different professors have made different choices: Some professors are worried about students' excessive use of ChatGPT Edu, adopting a class examination method using answer sheets and answer books, or using AI detection tools to identify AI-generated assignments. At the same time, some professors have incorporated ChatGPT Edu into their courses. They allow students to use AI to complete assignments and require the submission of chat records using AI.
But that hasn’t diminished confusion among faculty and students about the use of AI in courses. In an interview with the media, a CSU professor directly pointed out the nature of the contradiction: The original intention of introducing AI technology is to assist teaching or research work, but it is also equivalent to planting a hidden "bomb" - will AI replace the professor's teaching and research work?
Adopting AI has helped ChatGPT Edu elevate its position in the public education ecosystem, in part because the university is investing public funds in a private, for-profit company whose main product could, in turn, put many taxpayers out of work.
(Source: ChatGPT Edu)
Judging by the reaction, CSU’s AI initiative reflects a crisis of institutional identity and a debate over the very nature of public education. When students and faculty lack a clear understanding of the capabilities needed to cultivate the next generation of social builders and the AI economy four years from now, how can we start talking about how to cultivate them?
The protests didn’t stop AI from coming to campus, and CSU once again renewed a new agreement with OpenAI as the agreement was about to expire. In addition, ChatGPT Edu has signed relevant agreements with a number of universities, including the University of Oxford, the University of Texas at Austin, and the Wharton School of the University of Pennsylvania.
But the real unanswered question is: When universities become more and more like technology platforms, is this an upgrade of education, or is it making education lose its soul? Who is responsible for the results of this AI experiment?
References:
1.https://www.nytimes.com/2026/06/01/magazine/ai-university-college-california.html
2.https://openai.com/index/openai-and-the-csu-system/
3.https://www.govtech.com/education/higher-ed/san-jose-state-university-creates-ai-librarian-position
4. https://www.insidehighered.com/news/tech-innovation/artificial-intelligence/2026/03/27/faculty-push-back-against-openai-deals
5.https://www.kpbs.org/news/education/2026/05/01/cal-state-struck-a-deal-with-openai-some-students-and-faculty-refuse-to-use-it