Author: Wei Fei, Associate Researcher, Vice Dean of the Teacher Development Institute, East China Normal University
Generative AI is a model and technology that generates content based on user intent, and can be seen as a tool for content creation. It is capable of generating new text, images, music, and even videos based on user perspectives. Today, a plethora of creative intelligence tools have emerged, leading to new lifestyles and work modes, with their widespread application gradually affecting social transformation.
In the field of education, the integration of generative AI can not only enrich teaching content and methods but also innovate educational teaching models, bringing revolutionary changes to education. Against the backdrop of the national push for digital transformation, understanding, recognizing, and learning about generative AI has become a core element for teachers to enhance their digital literacy and implement core competency reforms.
However, what about the safety of generative AI? How can we use generative AI properly? The author will elaborate on four aspects: recognizing risks, understanding scenarios, acquiring skills, and innovating applications, to help teachers better meet new challenges and requirements, and become smart teachers in the AI era.
1 Recognizing Risks: Understanding the Working Principles of Generative AI and Its Potential Multiple Risks
Generative AI learns from large-scale datasets through understanding and deep analysis, mimicking human language and behavior patterns to generate content that meets user needs. Its power lies in the diversity and creativity of the content it generates; however, it also harbors numerous risks and challenges.
First is the risk of value orientation. Generative AI systems may absorb and amplify biases present in their training data during the training process, leading to generated content with erroneous guidance. Algorithmic bias and the information cocoon effect are two related and significant issues. If the training data contains erroneous ideological concepts, the content generated by AI may negatively impact users’ values and judgment.
Secondly, there is the risk of data security. Generative AI requires a large amount of data for learning and training, which may contain private and sensitive information, inevitably raising data security concerns. If the input student information is not adequately protected, it may lead to student privacy breaches. Additionally, a lack of clear data management standards could result in improper management and use of information, leading to data misuse or unauthorized use.
Finally, there is the risk of content quality. The essence of content generated by generative AI is pattern prediction, relying on massive amounts of high-quality data for learning and content generation. In the context of deep implementation of curriculum reform, if the data used for AI training (including policies, theories, and cases) is not comprehensive or of poor quality, the generated content may exhibit significant bias or uncertainty. For teachers, if they lack sufficient professional literacy and experience to reasonably evaluate and judge AI-generated content, it may lead to serious misguidance, negatively impacting students.
To address these risks and challenges, some higher education institutions and international organizations have begun to formulate guidelines for the use of generative artificial intelligence. For example, UNESCO released the “Recommendations on the Ethics of Artificial Intelligence” in 2021, and the “Guidelines for Generative Artificial Intelligence in Education and Research” in 2023, providing important learning resources for teachers. Teachers should strengthen their awareness of data ethics and security, actively engage in professional learning, enhance their professional literacy, and continuously improve their ethical and professional sensitivity and judgment in the use of generative AI, ensuring compliance with ethical safety and professional standards. This is the basic premise for conducting educational applications.
2 Understanding Scenarios: Familiarizing with Typical Educational Applications of Generative AI
Generative AI trains on large-scale datasets that include educational theories and practical cases, teaching methods and strategies, and online literature resources, to generate creative suggestions based on educational concepts and best practices, effectively assisting teachers in everyday practical applications. Typical applications of generative AI include: optimizing teaching design, assisting resource development, constructing inquiry environments, analyzing student data, enhancing management efficiency, and empowering research.
In teaching design, based on the analysis of teaching objectives and targets, teachers can use generative AI to assist in the design of key elements such as unit big concepts, higher-order questions, real-life contexts, learning scaffolding, and learning activities, to implement core competency cultivation and achieve high-quality education.
In resource development, AI can support the design and creation of presentations, images, and videos, enriching resource forms and improving resource development efficiency.
In constructing inquiry environments, AI can create high-interaction environments based on virtual reality, metaverse, and other technologies that promote higher-order thinking, enhancing students’ problem analysis, reflective evaluation, and critical thinking skills.
In student data analysis, AI can clean evaluation data, calculate key indicators, analyze data based on learning analytics models, and generate visual charts, allowing teachers to focus more on data understanding and student analysis.
In daily management, generative AI can help teachers record and analyze classroom data, assist in writing notifications, reports, and communication emails with parents, reducing the workload of administrative tasks and improving management efficiency.
In research, generative AI can assist in designing and refining research questions, hypotheses, data collection tools, and research outlines, standardizing report formats and enriching research perspectives.
3 Acquiring Skills: Mastering Key Skills for Operating and Using Generative AI
First, teachers must master prompt engineering, i.e., writing effective prompts. Generative AI requires a clear understanding of human-specific needs to provide appropriate feedback. In other words, high-quality input questions lead to higher quality responses. Generative AI prompts must accurately express human roles, work instructions, contexts, relevant data, and output requirements (including evaluation criteria for expected results). Work instructions specify the particular tasks that generative AI is expected to perform, such as designing teaching activities based on the 5E instructional model; the context provides background information related to the task, such as the target audience for the activity, duration, environmental information, and may include theoretical models, evaluation standards, reference examples, or learning materials to guide generative AI in better responding to task requirements.
Secondly, a process-oriented mindset is needed to articulate tasks in the language understood by large models. If the task instructions are overly complex, the questions unclear, or the requirements vague, generative AI may struggle to produce satisfactory results. When posing questions, it is essential to consciously break down tasks and processes, dividing complex tasks into multiple specific smaller tasks. For example, regarding the question “How to improve classroom teaching efficiency,” one can decompose the problem from the perspectives of classroom management, teaching methods, teaching content, and student evaluation, formulating sub-questions like “How to effectively manage student discipline in class?” “What teaching methods can enhance student engagement and interaction?” “How to design learning content based on students’ levels and interests?” “What methods can accurately assess learning outcomes?” to articulate needs.
Additionally, and most crucially, teachers must evaluate and judge the generated content. By continually enhancing their professional literacy and judgment, they should conduct detailed analyses and rational assessments of the accuracy of generative AI outputs, the implicit values, and the potential impacts on teaching and students. When faced with the possible “hallucinations” and “nonsense” of generative AI, teachers should possess a keen critical awareness, utilizing their professional knowledge, practical experience, and literature to comprehensively examine the outputs of generative AI. Through the “human-in-the-loop” collaborative mechanism, they should fully engage in the answer generation process, promptly identifying and correcting errors generated by AI, and actively constructing their understanding of the problem.
4 Innovative Applications: Exploring Innovative Applications of Generative AI in Deep Integration with Education
Teachers should innovate assignment and task design to enhance cognitive challenges. As “digital natives,” students will naturally use AI to assist their learning. The low threshold and boundless generative capabilities may foster students’ plagiarism and dependency behaviors, which are detrimental to problem-solving abilities and critical thinking development. To avoid learning inertia or the phenomenon of “having results without growth,” teachers must innovate task designs to ensure students have ample opportunities for cognitive training and growth. For instance, using real-life scenarios and open-ended tasks that promote problem-solving to increase cognitive challenge levels, guiding students to engage in process evaluations and reflections to develop self-awareness and monitoring abilities.
Teachers should actively research and accumulate technical resources related to generative AI to create autonomous learning and inquiry spaces for students. For example, creating virtual laboratories for students to conduct scientific experiments and inquiry activities in a safe, controlled environment; simulating various real-life scenarios to help students learn in specific contexts, enhancing the practicality and interactivity of learning; constructing data analysis environments that allow students to explore complex relationships between data, cultivating data analysis capabilities and critical thinking.
Teachers should strengthen interdisciplinary awareness in the application of generative AI to enhance students’ comprehensive qualities. Digital technologies, especially generative AI, inherently possess multi-disciplinary integration characteristics, becoming important tools for conducting interdisciplinary activities. Based on the scenario creation, inquiry support, and personalized guidance of generative AI, teachers can design and implement interdisciplinary projects. For example, using generative AI to create historical scenarios for immersive inquiry-based learning; in art courses, generative AI can generate creative works to stimulate students’ creativity. Through these interdisciplinary projects, students can not only better understand the intrinsic connections between subject knowledge but also develop innovation capabilities and critical thinking.
In summary, when facing generative AI, teachers need to maintain an open and cautious attitude, fully leveraging its advantages while being alert to its potential risks. Only by consistently adhering to the principle of prioritizing students’ physical and mental health and comprehensive development as the application standard, and balancing technological application with ethical safety, can we ensure that it truly empowers education. (Source: China Education News)
END
Submission Email:[email protected]. Articles published in the electronic journal “Henan Education Informatization” can be considered for the provincial Excellent Achievements Award in Education Informatization (for details, please refer to the “Notice from the Office of the Henan Provincial Department of Education on the 2023 Annual Application for the Excellent Achievements Award in Education Informatization”).
Disclaimer: This WeChat reposted article is for non-commercial educational and research purposes and does not imply endorsement of its views or verification of its content’s authenticity. If the reposted article involves copyright issues, please contact us immediately, and we will modify or delete related articles to protect your rights.