Lecturer Development Requirements to Teach AI Literacy: Lessons from a European Student Survey

Picture an AI Playground session at a university — a drop-in workshop where lecturers can explore different AI tools, ask questions, and experiment at their own pace. A trainer circulates the room, helping people try out ChatGPT, Claude, image generators, transcription tools. The atmosphere is curious, exploratory. As the session unfolds, something becomes clear: the lecturers in the room are at widely different levels. One has been using AI daily for months and wants to discuss detection tools for synthetic media. Another is opening Duck.ai for the first time and asks, “So this just… knows things?” A third is somewhere in between — aware that students are using these tools, but unsure what to tell them about verifying what AI produces.

The trainer realizes: before we can help lecturers teach AI literacy to students, we need a baseline. We are not there yet. What the trainer experienced in that room is not an anomaly. It reflects a structural gap that is, as this piece will argue, one of the most consequential yet least-addressed challenges in contemporary faculty development.

This is the challenge facing centers for teaching and learning, faculty development units, and anyone responsible for preparing lecturers for the classroom. We talk constantly about AI literacy for students — whether they should use ChatGPT for assignments, how to detect AI-generated text, what new citation practices we need (Ng et al., 2021). But we talk far less about a prior question: what do lecturers themselves need to learn before they can teach any of this?

I write this from my position as an instructional designer and researcher involved in METACOG, an Erasmus+ funded project focused on AI literacy and combating misinformation and disinformation in higher education. The project brings together partners from Finland, the Netherlands, Montenegro, and Slovakia. In early 2025, METACOG surveyed 360 students across these four countries to understand their current practices, skills, and gaps when it comes to recognizing and managing false information — particularly in an age of generative AI.

What we found should give any faculty developer pause.

Students Are Already There — Lecturers Are Catching Up

The survey data¹ gathered through the METACOG project offer a window into current student practice across four European countries. Nearly half of the students we surveyed — 44 percent — reported using conversational AI tools like ChatGPT or Claude as a source of information. Not just for writing assistance, but for information. Meanwhile, over three-quarters still rely on traditional search engines, and two-thirds turn to online platforms like YouTube, blogs, and social media. These are not exclusive categories; students layer their sources, often without clear strategies for reconciling conflicting information.

At the same time, students are aware that misinformation is a serious problem. Nearly 80 percent said they believe disinformation severely impacts society. And 95 percent reported having encountered false content while researching a topic — most of them occasionally, but a quarter of them frequently.

So students are using AI. They know misinformation exists. They have encountered it themselves. But are they equipped to identify it?

Here the picture becomes more complicated. When we asked students to rate their own verification practices, cross-referencing multiple sources was common — about 70 percent said they do this often or always. But when we asked about formal fact-checking tools like Snopes, FactCheck.org, or Google Fact Check Explorer, only 13 percent reported using them regularly. Over 40 percent said they never use such tools at all.

The gap widens further when it comes to AI-generated content specifically. Among all the skills we asked students to self-assess — identifying fake news, evaluating source credibility, using digital verification tools — the lowest-rated skill was detecting manipulative AI-generated content like deepfakes. This is precisely the area where the information environment is evolving fastest, and it is the area where students feel least prepared.

Lessons for Faculty Development

These findings point toward a set of professionalization goals that centers for teaching and learning — and anyone responsible for supporting lecturers — should be addressing. Five interrelated priorities emerge from this data, each addressing a different dimension of what AI literacy means in practice for the educators responsible for developing it in their students.

First, lecturers need to know which tools students are using. It is difficult to guide critical AI use if you do not know which tools your students rely on, or how they are integrating AI into their research and writing workflows. This is not about surveillance; it is about meeting students where they are. Faculty development should include structured opportunities for lecturers to explore the AI tools their students are using — not to police them, but to understand them.

Second, lecturers need to move from prohibition to pedagogy. Our survey found that students are overwhelmingly willing to use AI tools for information verification — over 75 percent expressed moderate to strong willingness. They are not resistant. But they lack guidance. Lecturers who simply forbid AI use, or who treat it solely as an academic integrity problem, miss an opportunity to teach critical engagement. Faculty development should help lecturers design assignments and classroom activities that teach students how to verify AI outputs, not just whether to trust them.

Third, lecturers need to learn verification as a structured skill set. Telling students to “check your sources” is insufficient. What does that mean in practice? Which tools? In what order? How do you triangulate between sources that contradict each other? Faculty development should introduce lecturers to concrete verification workflows — lateral reading (Wineburg & McGrew, 2019), reverse image searches, source tracing — and help them model these practices explicitly in their teaching.

Fourth, lecturers need technical literacy in AI-generated content. You cannot teach students to detect deepfakes if you do not understand how they are made. You cannot explain why large language models hallucinate citations if you do not know what a language model is (Ji et al., 2023). This does not mean every lecturer needs to become a machine learning expert. But a baseline understanding of how generative AI works — its affordances and its failure modes — should be part of professional development programming.

Fifth, AI literacy needs to be contextualized by discipline. Our survey included students from computer science, law, journalism, education, and other fields. Their open-ended responses made clear that they want guidance relevant to their own domains. One student wrote: “In the Law field, I believe that the only good sources are government pages, books, PDFs with law-related documents, or sites that are known to be reliable: EUR-Lex.” Faculty development should help lecturers translate AI literacy into their specific disciplinary contexts, not treat it as a generic add-on.

Taken together, these priorities point toward a vision of faculty development that is structured, discipline-sensitive, and technically grounded. Whether institutions are prepared to support such a vision — or whether they are even positioned to do so — is a different, and more unsettling, question.

The Institutional Tension

There is an uncomfortable reality beneath all of this. Many institutions are not ready.

In the Netherlands, where I work, some universities of applied sciences still do not provide institution-approved AI tools for lecturers and students to use. Lecturers are left to navigate commercial tools on their own, without clear guidance on privacy, data protection, or pedagogical integration. Students, meanwhile, are using these tools regardless. The gap between institutional policy and everyday practice creates confusion and undermines any coherent approach to AI literacy.

This is not unique to the Netherlands. Across Europe and beyond, institutions are struggling to keep pace with the speed of AI adoption. Reports from the European University Association on AI in higher education, or the 2023 UNESCO guidance on AI and education (Holmes et al., 2023), are publicly available and directly relevant. Faculty developers often find themselves preparing lecturers for a landscape that their own institutions have not yet officially sanctioned.

And yet the work cannot wait. What makes this institutional hesitation particularly striking is that students themselves have already articulated what they need — not in vague terms, but with specificity and creativity. When we asked students how higher education could better prepare them to manage disinformation, their suggestions were concrete: integrate media literacy across disciplines, offer hands-on workshops, teach specific verification tools, provide access to reliable source databases. One student proposed a creative approach: “Every semester there will be one workshop in which the lecturer believably provides false information. And before the end of the workshop the students have to determine that it was disinformation.”

Students are asking for this. They recognize the need. The question is whether institutions — and the faculty developers within them — will rise to meet it.

Conclusion

Back at the AI Playground, the trainer makes a note: next time, start with a baseline assessment. Not to gatekeep, but to differentiate. The lecturer who has never used ChatGPT and the lecturer who wants to discuss deepfake detection cannot be served by the same session. And both need something before they can help their students navigate AI-generated misinformation.

AI literacy is not a student problem with a lecturer solution. It is a shared challenge that requires lecturers to develop new competencies alongside their students. Centers for teaching and learning, faculty development units, and instructional designers have a critical role to play — not through one-time workshops, but through sustained professional learning that meets lecturers where they are and builds from there (Darling-Hammond et al., 2017).

The METACOG survey revealed students who are aware, willing, and asking for help. It also revealed gaps in verification practices and technical understanding that will not close on their own. If we want students to be literate enough to recognize AI-based misinformation, we first need lecturers who are literate enough to teach them.

That is not a criticism. It is a starting point.

References

Darling-Hammond, L., Hyler, M. E., & Gardner, M. (2017). Effective teacher professional development. Learning Policy Institute.

Holmes, W., Miao, F., & Giannini, S. (2023). Guidance for generative AI in education and research. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000386693

Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023). Survey of hallucination in natural language generation. ACM Computing Surveys, 55(12), 1–38. https://doi.org/10.1145/3571730

Ng, D. T. K., Leung, J. K. L., Chu, S. K. W., & Qiao, M. S. (2021). Conceptualizing AI literacy: An exploratory review. Computers and Education: Artificial Intelligence, 2, 100041. https://doi.org/10.1016/j.caeai.2021.100041

Wineburg, S., & McGrew, S. (2019). Lateral reading and the nature of expertise: Reading less and learning more when evaluating digital information. Teachers College Record, 121(11), 1–40.

About the Author

Tamara N. Lewis Arredondo, Ph.D. is a Senior Lecturer, Teacher Trainer, and Researcher with the Multilevel Regulation research group at The Hague Center for Teaching and Learning, The Hague University of Applied Sciences, the Netherlands. Her work focuses on the integration of AI in teaching and learning, global citizenship education, and inclusive practice in higher education. She is a contributor to the Erasmus+ funded METACOG project and the developer of the AI Playground at The Hague University of Applied Sciences.

Correspondence: tnlewis@hhs.nl

Cite this article

Lewis Arredondo, T. N. (2026). Lecturer development requirements to teach AI literacy: Lessons from a European student survey. Society and AI. https://societyandai.org/insights/lecturer-development-ai-literacy-european-student-survey/

Write for Society & AI

We welcome contributions from educators, researchers, policymakers, and practitioners examining the intersection of artificial intelligence, education, and society. All work published on Society & AI is made available free of cost and in open access, in service of education and the public good. If you have a perspective to share, send your proposal to the Editorial and Content Director at submissions@societyandai.org.