From Research to Practice: Guidelines for AI in the Creative Industries
Artificial intelligence is already changing how creative work is taught, produced and managed. Across the AICI project, our research has also demonstrated why responding to that change requires more than learning how to use the latest generation of AI tools.
Questions around AI increasingly intersect with education, professional practice, copyright, skills, creative agency, cultural representation and the future of work. The implications also look different depending on where you sit within the creative ecosystem. An educator preparing students for their first professional roles faces different decisions from a studio introducing AI into an established production pipeline, while policymakers must consider the wider conditions in which both education and industry operate.
Drawing together findings from across AICI, we have developed three sets of Guidelines and Recommendations for:
• Educational and training institutions
• Creative industry representatives
• Decision makers and policy stakeholders
Each publication translates the research into practical recommendations for its audience while contributing to a shared principle that runs throughout the project: the future of AI in the creative industries will be shaped as much by human judgement as by technological capability.
For Education: Preparing for AI Without Losing the Foundations
Generative AI is already part of students’ academic and professional environment. Our research therefore points towards structured and discipline-sensitive approaches to AI education, rather than ignoring its use or responding primarily through prohibition.
Student research found widespread AI use while also identifying concerns around over-reliance, academic integrity and unclear institutional boundaries. Importantly, students themselves largely favoured regulated rather than unrestricted use. At the same time, creative-industry teachers repeatedly emphasised the continued importance of fundamentals, including artistic judgement, design literacy, critical thinking, debugging, creative process and an understanding of craft.
The educational recommendations respond to this by placing AI readiness ahead of AI dependency.
They call for institutions to build critical AI literacy across curricula without structuring education around individual platforms; protect craft, creativity and human judgement; establish clear and discipline-specific expectations for AI use; and rethink assessment so that the learning and creative process remains visible.
The guidelines also identify educator development as a strategic priority. AI literacy cannot depend on individual teachers finding time to keep pace with technological change independently. Continuous professional development, responsible experimentation and stronger understanding of copyright, ethics, data, bias and attribution need institutional support.
Ultimately, preparing students for an AI-augmented creative sector means developing professionals who can evaluate new technologies rather than simply operate them. Entrepreneurship, collaboration, resilience, self-directed learning and adaptability become particularly important in a professional landscape where individual tools and workflows will continue to change.
Explore the full Guidelines and Recommendations for Educational and Training Institutions for all eight recommendations, supporting evidence and practical guidance.
For Industry: Focus on Effective Adoption
For creative industry representatives, the research raises a different question: where does AI actually create meaningful production value?
Generative AI offers opportunities for creative digital production, but the evidence does not support assuming that adoption will automatically reduce labour, production time or cost. AI currently appears most useful when augmenting professional expertise, automating selected tasks and supporting exploratory work, while skilled professionals remain central to artistic direction, contextual understanding, consistency, quality assurance and ethical decision-making.
This distinction becomes particularly important when evaluating efficiency. A tool may generate an asset quickly while creating substantial additional work through prompting, correction, supervision, rebuilding or integration into the wider pipeline. The recommendations therefore encourage organisations to begin with a clearly defined production problem and assess AI across the complete workflow.
Human responsibility is equally important. Where AI contributes to production, accountability for artistic direction, asset approval, narrative and cultural coherence, accuracy, copyright, provenance, representation and final quality control should remain clearly defined.
The industry guidelines consequently recommend controlled pilots before scaling adoption, clear internal governance around intellectual property and data, transparency with employees and creators, and continued investment in workforce skills alongside investment in technology.
Creative professionals should also be involved in decisions about which systems are adopted and where they are used. Those working closest to production bring the professional knowledge required to identify genuine efficiencies, limitations and risks.
The objective is effective AI adoption: introducing technology where it creates demonstrable value while maintaining creative quality, professional expertise, intellectual property, workforce trust and cultural integrity.
Explore the full Guidelines and Recommendations for Creative Industry Representatives for guidance on evaluating, governing and implementing AI within professional creative workflows.
For Policymakers: Creating the Conditions for Responsible Innovation
AI is reshaping creative production, education and labour simultaneously. For policymakers and decision makers, this means AI cannot be addressed solely as a technology issue.
The research identifies interconnected questions around skills, copyright, cultural diversity, workforce transition, sustainability, access to innovation, ethics and competitiveness. Legal uncertainty can affect how confidently creators, studios and educators engage with emerging technologies, while fragmented educational responses and gaps in educator preparedness risk widening the distance between technological development and workforce capacity.
The policy recommendations therefore focus on creating the conditions in which innovation and protection can reinforce one another.
This includes clearer but adaptable frameworks around authorship, copyright, training data, consent, licensing and ownership; common responsible-AI standards; and sustained investment in human capacity alongside technological infrastructure.
Access also matters. Shared AI and creative technology labs, innovation sandboxes, education–industry test environments and other collaborative infrastructure can give SMEs, independent creators and educational institutions opportunities to experiment without carrying the full financial and technical burden independently.
The recommendations also recognise cultural diversity as a strategic consideration. AI systems can reproduce biases and dominant cultural aesthetics present within their training data. Supporting culturally and linguistically diverse research, responsible datasets, representation assessment and culturally specific creative practices will be important if technological competitiveness is to develop without contributing to cultural homogenisation.
Workforce policy will similarly need to look beyond predictions about which occupations may disappear and examine how individual tasks and competency requirements are changing. Accessible reskilling and lifelong-learning pathways can help creative professionals move towards emerging and augmented roles as working practices evolve.
Finally, policy itself must remain responsive. Recurring evidence collection and sustained connections between education, industry, research and government can create a feedback loop in which changing professional practice informs education, research evaluates outcomes, policy responds to evidence and knowledge returns to the wider creative ecosystem.
The policy goal should therefore extend beyond accelerating AI adoption. Europe needs the conditions to adopt, question, shape and develop AI responsibly while maintaining creative agency, cultural diversity, professional sustainability and international competitiveness.
Explore the full Guidelines and Recommendations for Decision Makers and Policy Stakeholders for the complete policy framework and recommendations.
A Shared Direction
Although the recommendations address three different audiences, the research points towards a common direction.
Education cannot prepare future professionals without understanding how industry practice is changing. Industry cannot implement emerging technologies sustainably without skilled professionals, responsible governance and a strong education pipeline. Policymakers cannot develop effective frameworks without evidence from the educators, researchers, creators and organisations experiencing these changes in practice.
The relationship between these groups therefore matters as much as the actions of any one of them.
Across all three publications, human judgement remains central. AI literacy means understanding capability and limitation; innovation requires evidence as well as experimentation; and technological progress must be considered alongside creative agency, rights, skills, culture and professional responsibility.
These guidelines are therefore intended as a practical starting point rather than a fixed response to a rapidly changing technology. As AI continues to develop, so too will the questions facing education, industry and policy. Sustained dialogue, experimentation and evidence will be essential to ensuring that the creative industries can respond to that change critically, responsibly and creatively.
