The ‘deep learn’ framework: elevating AI literacy in higher education

The integration of artificial intelligence (AI) into higher education is transforming how students learn and educators teach. From giving personalised learning experiences to instant feedback, AI is opening new doors for enhanced educational experiences​​. 

Recent studies underscore the benefits of AI in education. A systematic review from 2016 to 2022 of 138 articles revealed significant advancements in AI applications in higher education. These applications included assessment, prediction, assistance, intelligent tutoring and managing student learning, emphasising AI’s transformative role​​. Scholars have reported on AI’s ability to adapt instruction to learner types, provide customised feedback, develop assessments and predict academic success​​.

However, this integration brings with it the concerns of de-skilling. As Gabi Reinmann’s 2023 discussion paper (in German) highlights, the term “de-skilling” refers to the potential erosion of human competencies, a subject that, while not extensively explored in the context of AI at universities, is of growing importance. As in sectors such as medicine and finance, where reliance on AI and automation has led to reduced hands-on skills and domain knowledge​​, there’s a risk in higher education, too. The extensive use of AI could diminish critical thinking, creative problem-solving and other essential skills. 

• Four steps for integrating generative AI in learning and teaching
• Teaching AI literacy: how to begin
• Universities, AI and the common good
• Advice spotlight: AI and assessment in higher education

Furthermore, ethical considerations, such as potential biases and privacy concerns in AI systems, add another layer of complexity to this rapidly evolving educational landscape​​. In light of these ethical challenges, how do we ensure that AI tools are used responsibly and inclusively in educational settings? 

Introducing the ‘deep learn’ framework

To navigate this AI-infused academic environment, I propose the “deep learn” framework. The acronym stands for: discover, engage, evaluate, probe, link, expand, adapt, reflect, navigate. It’s more than a set of guidelines; it’s a mindset for engaging with AI in a thoughtful, critical and ethical manner. Each element has a learning objective and practical exercises for students. 

Discover: encourage exploration

Objective: Students use AI to uncover areas of study, research and interest.

Application: For example, students might use AI to explore emerging trends in their fields, gather diverse perspectives or uncover historical data relevant to their studies.

Engage: promote meaningful interaction

Objective: Move beyond passive queries and actively engage with AI as a learning partner.

Application: Students could use AI to simulate real-world scenarios or participate in AI-driven discussions on complex topics.

Evaluate: foster critical assessment

Objective: Students can critically assess AI-generated information for accuracy, bias and relevance.

Application: Students could compare AI responses with established research to identify discrepancies or confirmations, enhancing their critical-thinking skills.

Probe: encourage in-depth enquiry

Objective: Students understand the need to ask follow-up questions and delve more deeply into subjects to gain a comprehensive understanding.

Application: Students can use AI to explore facets of a topic, asking iterative questions based on initial AI responses to gain deeper insights.

Link: connect AI insights to academic concepts

Objective: Highlight the importance of integrating AI insights with the academic. 

Application: Students can draw parallels between AI-generated information and theoretical concepts from their coursework, fostering a holistic understanding and gaining insight into practical implications.

Expand: broaden intellectual horizons

Objective: Students can use AI to explore perspectives and ideas beyond their current understanding.

Application: Students might use AI to generate creative solutions for a project or to understand global perspectives on local issues. For example, students could use AI to analyse discussion forums and social media from different cultures to enhance a multicultural curriculum.

Adapt: cultivate flexibility

Objective: Students develop adaptability in approach and thought, and the ability to adjust to the dynamic nature of AI.

Application: Students learn to refine their queries based on AI responses, customising their learning approaches as needed.

Reflect: instil thoughtful consideration

Objective: Students can reflect on the broader implications, ethics and impacts of AI-generated information.

Application: After interacting with AI (for example, to analyse case studies in the use of AI for hiring), students could reflect on how AI’s insights might affect societal issues, ethical considerations or future trends.

Navigate: master AI’s capabilities and limitations

Objective: Students can work within AI’s strengths and weaknesses, using it to its fullest potential.

Application: Students should learn to identify when AI is most useful and when human judgement is paramount, balancing AI’s capabilities with its limitations. 

The framework sets a foundation for students to interact with AI in a multifaceted and conscientious manner, addressing both the immense potential and the inherent challenges of AI in education. We must also envision a path that not only embraces these principles but also actively shapes the role of AI in enhancing educational outcomes and maintaining essential human skills.

The way forward for AI in higher education

Acknowledging these challenges paves the way for a balanced approach to AI’s role in higher education. The focus should shift towards a model that harnesses AI’s potential to enrich learning experiences, while also actively countering the risks of de-skilling. This necessitates using AI in a way that complements and enhances human abilities (rather than supplants them). We need a collaborative relationship between human and machine intelligence​​. 

In practical terms, this translates into educational settings where AI is leveraged for personalised learning journeys, simultaneously fostering active, critical engagement from learners. Educators and institutions must prioritise skills that AI cannot replicate, such as empathy, ethical judgement and nuanced problem-solving. Policymakers and educational leaders are called upon to develop frameworks and policies that support ethical AI integration, promote digital literacy and safeguard against the erosion of essential human skills. 

Additionally, ongoing research should aim to understand the long-term impacts of AI in education, exploring effective hybrid teaching models and developing AI tools that are transparent, equitable and culturally sensitive. By steering AI integration with a thoughtful, human-centric perspective, higher education can leverage AI’s strengths while preserving and nurturing the skills that are quintessential to human intellect. This balanced strategy ensures that AI becomes a powerful ally in the realm of education, fortifying rather than replacing the critical human skills and competencies at its core.

Birgit Phillips is director of academic professional development and AI at FH Joanneum University of Applied Sciences in Graz, Austria, and founder of Phillips Learning Lab. 

If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.