³Ô¹ÏÍø

Provost and Executive Vice ³Ô¹ÏÍø for Academic Affairs Ann Stevens

Ìý

I would like to begin by acknowledging the reactions many of you have had to theÌýCU system’s recently announced partnership agreementÌýwith OpenAI to provide ChatGPT Edu access to all CU students, faculty and staff across all four campuses. I have heard concerns about privacy, academic integrity, corporate influence, sustainability and whether this decision reflects who we are and how we teach, learn and conduct research at CU Boulder. These questions matter, and they reflect the care this community brings to our academic mission and to shared governance.

I would also like to be clear about what this agreement is and what it is not. This agreement does not require the use of generative AI in classrooms or research, nor does it diminish faculty authority over pedagogy, curriculum or assessment. It does not replace existing tools or limit future choices. Instead, it provides a secure, institutionally supported option for a technology many in our community are already encountering and using, often without the protections we would want to have in place.Ìý

Our current data show that more than 28,000 users on campus already have registered ChatGPT accounts using their @colorado.edu credentials, including more than 3,000 faculty and staff. Although that statistic is limited to users of CU email credentials, countless other users access tools like ChatGPT for work or studying using their personal email addresses as well.

Generative AI is now part of the world our students will enter and the environment in which scholarship, teaching and administrative work take place. Choosing not to engage would not shield us from these changes. Our responsibility is to engage thoughtfully by setting clear expectations for use, providing AI literacy education, safeguarding academic freedom and data privacy, encouraging open inquiry and dialogue, and ensuring access does not depend on personal resources or departmental budgets.

This agreement is a starting point not a final answer. It is intentionally time limited and will be evaluated and shaped by input from faculty, staff and students across disciplines. Above all, it reflects a commitment to move forward with care and intention, guided by our values and informed by our community.

We will not get every aspect right immediately. A transparent, time‑limited approach allows us to learn together, reduce risk and make thoughtful decisions about our future while protecting our academic mission.

Below, I’ve addressed several of the most common questions I’ve heard. I also encourage you to take a look atÌýCU Boulder’s ChatGPT Edu FAQs for additional information.

Does this agreement require faculty or students to use generative AI in teaching, learning or research?

No. This agreement does not require or expect faculty or students to use generative AI. Decisions about course design, assignments, assessment and research remain with faculty, departments and programs. Academic standards and expectations around original work are unchanged.Ìý

At CU Boulder, we recognize that what is appropriate in one discipline may not be appropriate in another, and that flexibility is essential to academic freedom.

Why enter into an agreement with OpenAI when concerns exist about corporate influence, vendor lock‑in and the broader AI ecosystem?

These concerns are real, and they are part of why this agreement is time limited and evaluative rather than open ended. Generative AI tools, particularly ChatGPT, are already widely used across higher education, including at CU Boulder, often on public platforms with little support or guidance and fewer data protections.

This agreement allows us to engage with a widely used tool on our terms, with governance, protections and the ability to reassess. It does not preclude other tools, future partnerships or alternative models, nor does it signal endorsement of a single vendor as a long‑term solution. It is a pragmatic step to reduce risk now while we continue broader conversations about innovation, competition and choice.

How does this agreement protect privacy, data security and academic integrity?

Data stewardship was a primary consideration. Under this agreement, your conversations are private to you by default, cannot be monitored by CU IT and are not used to train OpenAI’s models. This represents a significant improvement over the risks associated with public versions of the tool.

The agreement also operates within existing CU policies on data classification, privacy and responsible use. Academic integrity expectations remain unchanged, and guidance and training will accompany rollout to support appropriate and transparent use. While no approach eliminates all risk, this substantially reduces exposure compared to the status quo.

How does CU Boulder ensure trust and shared responsibility if legal requests involve OpenAI and university data?

Unless otherwise prohibited by law, if any request for university data through our OpenAI tool is made to OpenAI, the university, by contract, will be informed and consulted from the outset. OpenAI works in partnership with CU Boulder to respond in a way that respects the university’s values, legal obligations and commitment to protecting its community. This collaborative approach ensures transparency, coordination and a shared commitment to acting in the best interests of CU Boulder’s faculty, staff and students to the fullest extent allowed by law.

How will the campus community be involved in shaping how this tool is used and evaluated?

Listening to the many voices in our campus community is essential. And, here, I want to acknowledge concerns that faculty, staff and students were not broadly consulted before this system‑level contract for ChatGPT was finalized. Shared governance is central to how we shape academic priorities at CU Boulder, and faculty, staff and student involvement will be critical as decisions about implementation, evaluation and future direction are made. This contract is not the end of the conversation. It is the beginning of a more structured and inclusive one.

As implementation unfolds, we are committed to engaging our campus community across disciplines and campus offices in evaluation, guidance development and future decision‑making about how, and whether, this tool continues to serve our academic mission. That includes listening carefully to concerns, critiques and enthusiasm.

Faculty involvement will also help ensure that investments in AI are matched by continued support for the intellectual work that defines this university: teaching students to question, contextualize and use emerging technologies responsibly.

CU Boulder’s strength has always been its ability to balance innovation with careful scrutiny. That is the standard we will continue to apply.

How long is this agreement, where is the funding coming from and what will it ultimately cost CU Boulder?

The agreement is a three-year, system‑level contract, renewable annually. The approximately $2 million cost is covered at the CU system level for the first year, not through individual campus, departmental or course budgets.

What it ultimately costs CU Boulder will depend on how the agreement is evaluated and whether it continues beyond the initial term. This agreement does not commit the campus to ongoing or expanded spending. Any future decisions will be informed by evidence of academic value, faculty input and shared governance processes.