# Generative AI Literacy Is Quietly Rewiring How University Professors Teach

By Rafiqul Islam Rabbi · AI · Published Thu, 30 Jul 2026 07:34:28 GMT · Updated Thu, 30 Jul 2026 13:34:28 GMT
Source: The Current Tribune — https://currenttribune.com/article/generative-ai-literacy-tpack-teaching

Generative AI is everywhere on campus, but that doesn’t mean it’s actually changing how professors teach. A new study of university teachers in Shandong Province, China, argues that the missing ingredient isn’t more tools — it’s generative AI literacy and a specific kind of tech-pedagogy expertise known as TPACK.

## How AI support really reaches the classroom

The research zeroes in on a simple question: when a university spends heavily on AI platforms, training, and digital infrastructure, does that support actually show up as better AI-enhanced teaching practice in the classroom?

To probe that, the authors surveyed 578 university teachers using a structured online questionnaire. They analyzed the responses with descriptive statistics, correlation tests, confirmatory factor analysis, and structural equation modeling — the toolkit you reach for when you want to understand not just if things are related, but how they influence one another.

The focus was on four pieces of the AI-in-education puzzle:

- **AI support** – the institutional backing teachers receive, from smart teaching platforms to formal training and access to AI resources.

- **Generative AI literacy** – teachers’ ability to understand, use, and critically evaluate generative AI in an educational context.

- **TPACK** – short for Technological Pedagogical Content Knowledge, a framework describing how teachers blend technology, pedagogy, and subject content in their practice.

- **Teaching practice** – what teachers actually do in their courses, especially around technology-integrated and AI-supported teaching.

The headline result: all four variables were significantly and positively correlated. More AI support is associated with higher generative AI literacy, stronger TPACK, and richer teaching practice. But the real story sits in how those links are chained together.

## The chain: from AI support to AI teaching practice

The study finds that AI support doesn’t just influence teaching practice directly. It also works through two powerful mediators: generative AI literacy and TPACK.

In other words, institutional support helps teachers build generative AI literacy; that literacy then feeds into their TPACK; and together, those capabilities show up as more meaningful AI-supported teaching practice. The researchers describe these as “chain-mediating” roles, because the impact flows step by step along that path.

They report several clear effects:

- AI support significantly boosts generative AI literacy.

- AI support also directly improves teaching practice, even before you factor in literacy and TPACK.

- Generative AI literacy significantly strengthens teachers’ TPACK.

- TPACK, in turn, significantly improves teaching practice.

The takeaway: just dropping AI tools into a university’s tech stack isn’t enough. When support is paired with deliberate efforts to build generative AI literacy, it strengthens the underlying TPACK muscles that determine whether AI is bolted onto existing lessons or genuinely integrated into how teaching happens.

## China’s AI push meets classroom reality

The study is grounded in a context where the AI rhetoric is already loud. In China, higher education policy explicitly encourages universities to embed artificial intelligence into curriculum design, classroom instruction, and teacher development.

In Shandong Province, universities have been building out digital infrastructure, deploying smart teaching platforms, and rolling out specialized AI training programs. On paper, it looks like a model environment for AI-enabled teaching.

But the survey data underscores a gap many institutions will recognize: teachers are using AI, yet mostly to make their own work more efficient — not to rethink teaching practice itself. Think generating slides faster, drafting feedback, or handling routine admin, rather than redesigning assessments, experimenting with AI-driven tutoring, or personalizing learning pathways.

The authors point out that teachers’ AI literacy is often in the “moderate to upper” range overall, but that hides big differences in depth. Some have strong theoretical understanding and practical skills; others struggle with advanced use or with ethical and pedagogical questions. Access to sustained training, real organizational backing, and intrinsic motivation emerges as central to closing those gaps.

## Why generative AI literacy is different

It’s tempting to see generative AI literacy as just another digital skillset, but the study hints at why it’s more complicated. Traditional ICT tools in education — slides, learning management systems, basic analytics — are mostly static or transactional. Generative AI systems, by contrast, are interactive, adaptive, and context-aware.

That means the literacy bar is higher. Effective use isn’t just about knowing which button to press. It’s about understanding what kinds of prompts elicit useful responses, how to check AI-generated content for accuracy and bias, when to encourage students to use AI, and when to restrict it.

This is where the TPACK link matters. Generative AI literacy on its own risks turning into novelty tricks. When it’s coupled with TPACK, teachers aren’t just asking, “What can this model do?” They’re asking, “How does this tool intersect with my discipline, my learning goals, and my assessment design?”

The study’s structural model suggests that as teachers’ generative AI literacy grows, it naturally feeds into richer TPACK — new ways of blending content, pedagogy, and AI tools into coherent course designs. That, more than any single platform, is what actually changes teaching.

![University faculty training session focused on generative AI literacy and TPACK for teaching](/media/2026/07/generative-ai-literacy-tpack-teaching-inline.webp)
*Generative AI literacy workshops are becoming the bridge between campus AI investment and actual teaching change. (Photo: European Committee of the Regions / BY-NC-SA via Openverse)*

## From tool adoption to teaching transformation

The findings also slot into a broader story about how technology adoption works in universities. Previous work has often focused on individual attitudes: self-efficacy, technology acceptance, beliefs about teaching, or personal readiness. Those matter, but they mostly capture whether teachers feel willing and confident to try new tools.

This study looks at a different slice of the process: a development pathway that runs from external institutional support through personal competencies and on to observable teaching behaviors. It treats AI support, generative AI literacy, TPACK, and teaching practice as successive stages in teacher professional development under AI.

That framing is quietly provocative. It suggests universities can’t simply train for “usage” and call it a day. If the goal is to see AI embedded into subject-specific teaching — not just parked in admin workflows — then AI support has to be designed as a pipeline:

- First, ensure robust, reliable AI infrastructure and clear access.

- Second, build generative AI literacy that covers theory, practical skills, and ethics.

- Third, explicitly connect AI training to TPACK, so teachers learn to weave AI into pedagogy and content, not treat it as an add-on.

- Finally, support the translation of these competencies into concrete teaching practice, from course redesign and assessment strategies to classroom interaction.

Crucially, the study’s data show that each link in this chain matters. Skip the literacy stage and AI support loses much of its punch in the classroom. Ignore TPACK and you end up with scattered experiments instead of systemic change.

## What university leaders should actually do

For higher education administrators, the study offers more than theory; it’s a rough checklist for making AI investments pay off.

First, treat generative AI literacy as a core professional competence, not an optional workshop. The research ties it directly to both TPACK and teaching practice, which means it’s effectively a lever for improving technology-integrated teaching across the board.

Second, redesign AI training away from tool demos and toward teaching problems. Sessions should start from real scenarios — designing AI-resilient assessments, supporting struggling students with AI tutoring, or managing academic integrity when students use generative tools — and then map back to the capabilities teachers need.

Third, embed TPACK thinking inside AI initiatives. Rather than separate “tech training” and “pedagogy workshops,” build programs where subject experts, instructional designers, and technologists co-develop AI-enhanced courses. That’s how you translate literacy into practice.

Finally, recognize that support must be sustained. The study points to ongoing training opportunities, organizational backing, and motivation as critical factors in lifting AI literacy beyond a baseline. Short, one-off sessions will not move the structural equation.

## What This Means

The research extends the TPACK framework into a generative AI era and backs it up with empirical data from hundreds of university teachers. The core message is blunt: AI support on its own won’t transform teaching. It’s the chain — AI support feeding generative AI literacy, which strengthens TPACK, which then reshapes teaching practice — that really matters.

For universities racing to keep up with generative AI, that should be a relief and a warning. You don’t need the flashiest model on campus. You need a strategy that treats teachers’ AI literacy and TPACK as the main infrastructure. If those are strong, the tools will follow. If they’re weak, no amount of AI branding will show up where it counts: in the classroom.

*Photo: European Committee of the Regions / BY-NC-SA via Openverse*