Fashion schools need their own AI literacy, review argues

Fashion higher education is facing a structural AI literacy gap that generic upskilling cannot close, according to a review published in Discover Artificial Intelligence. The paper, authored by Srikant Manchiraju and published open access on 23 October 2026, argues that simply bolting AI tools onto existing fashion curricula misunderstands the problem. Drawing on a critical integrative review of interdisciplinary scholarship, it proposes a new conceptual framework, the Fashion AI Literacy Framework (FALF), built specifically around the creative, ethical, commercial and cultural terrain of the fashion world. The framework is presented explicitly as a conceptual proposition that still requires empirical validation through curriculum pilots and outcome studies, but the evidence base behind the underlying gap, the author argues, is already substantial.

The commercial backdrop explains the urgency. In 2023, the luxury conglomerate LVMH became a corporate affiliate of Stanford’s Human-Centered AI institute, Prada deployed AI in a fragrance campaign, H&M used generative models to produce digital clothing collections, and New York Fashion Week hosted its first runway shows explicitly engaging with AI-generated design concepts. By 2025, industry analysts were projecting that the AI fashion market would exceed $60 billion by 2034. Yet the same industry analysis revealed a widening gap between ambition and readiness: a large majority of fashion companies planned to increase their generative-AI investment, while only a small fraction considered themselves ready to deploy it at scale, with workforce training cited as a leading barrier to adoption.

The review employed a critical integrative methodology, layering integrative review practice onto the transparency principles of established scoping review guidance, including the Joanna Briggs Institute framework and PRISMA-ScR reporting principles where applicable. Structured searches across Scopus, Web of Science, ERIC, ACM Digital Library, IEEE Xplore, Google Scholar and Emerald Insight covered three conceptual blocks: AI literacy, fashion and creative education, and higher education frameworks. Because the intersection of AI literacy and fashion education is almost entirely absent from the literature, the author ran the blocks in separate pairings rather than only as a three-way combination, capturing more than 90 sources published between 2018 and February 2026. The author acknowledged several limitations: non-English publications were excluded, no formal quality appraisal of included studies was conducted, record-level screening counts were not logged prospectively so no PRISMA flow diagram applies, and screening was performed by a single author without a second screener or inter-rater reliability.

A central complication the review maps is that AI literacy itself remains a contested concept. An integrative review of 124 AI literacy studies cited in the paper found that, despite the topic’s prominence since generative AI, its definition remains vague and there is little consensus on how to design interventions. The dominant reference point remains Long and Magerko’s definition of AI literacy as a set of competencies enabling individuals to critically evaluate AI technologies, communicate and collaborate effectively with AI, and use AI as a tool at work and home, operationalized through 17 core competencies grouped into clusters such as recognizing AI, understanding how it works, critically appraising it, using it and perceiving its social and ethical implications. Subsequent work has added refinements: Ng and colleagues proposed four dimensions spanning knowing, using, evaluating and ethics, while later scholars introduced metacognition, self-reflection and collaborative mindsets as essential components.

The review surveys the major frameworks governing the field and finds none built for creative fashion disciplines. Long and Magerko’s framework predates the generative AI revolution. Ng and colleagues’ four-part structure places ethics at its center but lacks granularity for disciplinary adaptation. Ismailov and colleagues distinguish AI literacy, meaning knowing, from AI competency, meaning applying with confidence, and foreground affective and self-reflective dimensions. Ahn and colleagues, synthesizing 29 studies, produced eight competencies and 18 sub-competencies differentiated by learner level, finding that higher education should shift from basic device use toward understanding data and algorithms and career-related competencies. The AI-TPACK framework adds AI-technological knowledge to the established TPACK model, explicitly theorizing the relationship between subject expertise and technological competency, which the author notes is particularly valuable for studio-based design education.

The stakes for the industry itself are concrete. Generative tools such as Midjourney, DALL-E, Adobe Firefly and Stable Diffusion have altered early-stage design, and systems employing generative adversarial networks can reconstruct, alter and expand fashion image elements, shifting the designer’s role from sole creator to co-creative sense-maker. One cited study noted that AI tools can slash design timelines from weeks to days, and LVMH, Zara and Burberry have publicly integrated AI into design development workflows. Machine learning now performs trend forecasting on social media data, sales histories and search patterns, identifying micro-trends weeks or months before human observers. AI-enabled 3D design software such as CLO3D and Browzwear reduces physical sampling, cutting material waste, while AI-driven demand forecasting targets fashion’s chronic overproduction. Recommendation systems, virtual try-on and dynamic pricing raise questions about data privacy, consumer autonomy and behavioral manipulation.

The workforce evidence reviewed is pointed. Kohli’s 2024 study found that skills imparted through the current curriculum are insufficient to work in the digital world and called for fundamental revision of fashion design bachelor’s programs. A qualitative study by Krasley and colleagues involving 21 participants across five countries found that AI digitalization is creating radical demand for intellectual, innovative and creative skills, and that industry and academia must collaborate in forecasting future skills requirements of the apparel workforce. A LinkedIn job posting analysis by Espejo Mendez and colleagues confirmed accelerating demand for AI-related skills in fashion alongside cautious, uneven adoption and significant knowledge gaps at both hiring and practitioner levels. Roberts and Bailey observed that fashion-related higher education programs have been late to adapt their curricula to these new technological standards. The World Economic Forum has estimated that 65 percent of students entering university today will eventually work in roles that do not yet exist, and the digital skills gap in fashion has been described not as a gap but as a shallow talent puddle.

Against this backdrop, the review advances five interconnected grounds for fashion-specific AI literacy. First, fashion AI tools carry domain-embedded limitations: students report that generative AI does not seem to understand much fashion design jargon, and instructors confirm that foundational design knowledge and clear conceptual expression are prerequisites for effective use. Second, the ethical stakes are structurally distinct, shaped by histories of racial exclusion, body normativity and cultural appropriation; one cited study documents AI-perpetuated formative bias, while empirical work by Solyst and colleagues demonstrated that AI systems prohibited Black Muslim teenage girls from generating outputs reflecting their identities, bodies and cultures, failures invisible to a general AI ethics curriculum. Third, studio pedagogy creates a unique epistemological tension, with instructors warning that AI threatens the hand-to-mind dialectic foundational to design cognition. Fourth, the professional heterogeneity of fashion graduates across design, merchandising, sustainability and communications demands modular, role-sensitive literacy. Fifth, fashion’s function as a cultural system mediating identity, power and desire requires engagement with postcolonial critique and feminist design theory.

FALF itself is organized around four developmental dimensions, Know, Use, Evaluate and Create Change, and seven competency domains. The dimensions describe increasing sophistication rather than lockstep stages, and the review stresses the scaffold is developmental in emphasis but iterative in practice. Two domains, AI conceptual literacy and fashion AI ethics, form a common core for all fashion students. The remaining five, fashion-AI tool proficiency, prompt craft and creative agency, algorithmic critique and representational awareness, data and trend literacy, and sustainable and inclusive AI practice, are modulated in emphasis across design, merchandising, communications, sustainability and retail-analytics pathways. The domains cluster into functional, applied and critical-ethical groups mapped onto established literacy traditions, with prompt literacy framed as the ability to communicate with AI systems using fashion-specific vocabulary and aesthetic intentions. Illustrative learning outcomes and assessment criteria are offered as proposals rather than validated instruments.

The paper is careful to separate what is documented from what is proposed. The literacy gap is grounded in the reviewed literature; FALF awaits empirical validation through curriculum pilots, student and instructor surveys, and longitudinal outcome assessment, and the author frames it as an opening provocation rather than a final answer. The conclusion argues that what is at stake exceeds curriculum design: fashion is one of the world’s most powerful cultural industries, shaping how billions of people understand beauty, identity and belonging. If AI is reshaping fashion, and the reviewed evidence indicates it already is, then the professionals guiding that reshaping, the paper contends, need a form of AI literacy that is conceptually grounded, critically oriented and ethically attuned to the discipline they serve.

Subject of Research: A conceptual framework for fashion-specific artificial intelligence literacy in higher education

Article Title: A fashion specific framework for artificial intelligence literacy in higher education

Article References: Manchiraju, S. (2026). A fashion specific framework for artificial intelligence literacy in higher education. Discover Artificial Intelligence, 6(1), Article 1609. https://doi.org/10.1007/s44163-026-02610-4

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02610-4

Keywords: AI literacy, fashion education, higher education, generative AI, Fashion AI Literacy Framework, curriculum design, integrative review, algorithmic bias, fashion industry, design pedagogy, ethical AI, workforce skills