Generative AI Literacy: A Comprehensive Framework for Literacy and Responsible Use

2025-04-26

AI Summary

Existing AI literacy frameworks inadequately address the opacity, hallucination propensity, and interactive complexity inherent in generative AI systems (e.g., ChatGPT, DALL·E), resulting in users’ diminished capacity to critically evaluate outputs and deploy such tools responsibly. To bridge this gap, this study systematically defines generative AI–specific literacy dimensions and proposes a 12-item structured framework encompassing prompt engineering, human-AI collaboration, ethical risk identification, and foundational model awareness. Drawing on interdisciplinary perspectives from education, human-computer interaction, and AI ethics, the framework integrates empirical user studies with expert consensus to yield the first internationally applicable, model-agnostic generative AI literacy standard. Validated across university curricula and corporate training programs, it demonstrably enhances users’ prompt quality, critical evaluation of outputs, and accountability—thereby establishing a foundational benchmark for civic literacy in the generative AI era.

Technology Category

• Humans and AI: AI for Accessibility
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness
• Natural Language Processing: Prompt Engineering / Prompting

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Agentic search

Abstract

After the release of several AI literacy guidelines, the rapid rise and widespread adoption of generative AI, such as ChatGPT, Dall E, and Deepseek, have transformed our lives. Unlike traditional AI algorithms (e.g., convolutional neural networks, semantic networks, classifiers) captured in existing AI literacy frameworks, generative AI exhibits distinct and more nuanced characteristics. However, a lack of robust generative AI literacy is hindering individuals ability to evaluate critically and use these models effectively and responsibly. To address this gap, we propose a set of guidelines with 12 items for generative AI literacy, organized into four key aspects: (1) Guidelines for Generative AI Tool Selection and Prompting, (2) Guidelines for Understanding Interaction with Generative AI, (3) Guidelines for Understanding Interaction with Generative AI, and (4) Guidelines for High Level Understanding of Generative AI. These guidelines aim to support schools, companies, educators, and organizations in developing frameworks that empower their members, such as students, employees, and stakeholders, to use generative AI in an efficient, ethical, and informed way.

Problem

• Lack of robust literacy for generative AI's unique characteristics
• Need guidelines to evaluate and use generative AI responsibly
• Framework required for ethical and informed generative AI usage

Innovation

• Proposes 12-item guidelines for generative AI literacy
• Organizes guidelines into four key aspects
• Supports efficient, ethical, and informed AI use