Maine's public university system is moving forward with a $1.39 million, two-year contract to bring ChatGPT Edu to its campuses, making generative AI tools available to roughly 25,200 students and 5,600 employees starting in July. The decision came after a competitive bidding process that ultimately selected OpenAI's education-focused chatbot over Google's Gemini, despite an earlier working group recommendation favoring Gemini.

The University of Maine System's choice reflects a broader shift in higher education: nearly 60% of U.S. college students now use AI tools in their coursework on a weekly basis, and about 1 in 5 use them daily, according to a 2026 study by the Lumina Foundation and Gallup. Rather than let students rely on free versions of ChatGPT, which can expose sensitive academic material to public training datasets, the system decided to invest in a controlled, institutional platform.

Why Did Maine Choose ChatGPT Edu Over Google Gemini?

The decision surprised some observers, since the university system's own working group had recommended Google Gemini as the primary provider in a report published last fall. That recommendation was based on Gemini's "favorable data security agreements and straightforward integration with existing UMS information technology systems," according to the report. However, when the system opened the bidding process to competing vendors, OpenAI's offer, submitted through vendor Carasoft, came in at roughly $600,000 lower than its competitors.

The cost advantage proved decisive. The system is currently in a five-day appeals period where other bidders can clarify their proposals, but they cannot submit new offers. Once that window closes, the contract will likely move forward with OpenAI.

"It's not coming with a mandate. It's really just recognizing that there's heavy usage out there. So, if we're going to have that kind of usage, let's do it in a safe space," said Ryan Low, the system's vice chancellor for finance and strategic AI integration.

Ryan Low, Vice Chancellor for Finance and Strategic AI Integration, University of Maine System

What Privacy Protections Does ChatGPT Edu Offer?

One of the central concerns driving the Maine system's decision was data security. Free versions of ChatGPT use user inputs to train and improve OpenAI's models, meaning that when students upload course materials, lecture notes, or assignments, that content can be incorporated into the company's broader AI training pipeline. ChatGPT Edu, by contrast, includes "robust" security and privacy controls that prevent prompts from being used for model training.

This distinction matters significantly in an academic environment. Low explained the risk of relying on free tools: "You can imagine a scenario where we've got thousands of students across the system using free versions of ChatGPT, constantly uploading course material and lectures, and now all of that material is going out to the free world. We don't want that". The ChatGPT Edu platform also allows users to build custom versions of ChatGPT for internal use and provides higher message limits than the free tier.

Low

How Will Maine Universities Implement ChatGPT Edu?

The rollout will be gradual and voluntary. The system plans to launch the tool on July 1, making it available through each person's university account. Rather than imposing top-down mandates about how the tool should be used, the Maine system is taking a decentralized approach that empowers individual departments and instructors to decide whether and how to integrate ChatGPT Edu into their work.

• Professional Development: The system plans to offer optional training opportunities for employees and students during the coming school year to help them use the platform effectively.
• Flexible Adoption: Decisions about how ChatGPT Edu is used in classrooms or other campus settings will be made by the people doing that work, though systemwide information policies protecting student privacy will still apply.
• Funding Review: The Board of Trustees will revisit funding for the second year of the contract as part of the next fiscal year budget process, with any student fees or additional costs subject to transparent public review.

The contract will be backed by investment income, and no additional fees will be charged to students during the initial two-year period.

What Led Maine Universities to Pursue a Systemwide AI Tool?

The decision to adopt a unified AI platform emerged from a recognition that AI tools are becoming increasingly common in the workplace. Individual departments and universities within the Maine system had already begun adopting various AI tools on a case-by-case basis, but this fragmented approach created inequities: only departments with sufficient budgets could afford to purchase AI software.

To address this gap, the university system established a working group in 2024 to explore how to integrate AI across its campuses. That group organized hands-on demonstrations of AI tools, including Gemini, and published a report recommending that the system adopt a flexible AI policy framework, make training and resources widely available, and ensure human oversight. The working group also noted that the system was already piloting AI for administrative tasks, such as processing transcript data and interpreting academic details like coursework requirements, GPAs, and credit transfers for the admissions review process.

The shift to a systemwide platform represents an effort to democratize access to AI tools across the university system while maintaining institutional control over data security and privacy. As the system noted in its announcement to employees and students, the goal is to "encourage employees to explore" AI tools "in a safe manner" rather than leaving them to rely on free versions that may compromise institutional data.

The Maine system's experience reflects a broader trend in higher education, where institutions are grappling with how to provide students with practical experience using AI tools that are becoming essential workplace skills, while also protecting sensitive academic and personal information from being absorbed into public AI training datasets.