https://orcid.org/0000-0001-9037-7626
Figures
Abstract
This paper applies the Policy Modeling Consistency (PMC) index model to quantitatively evaluate 32 Chinese policies (2016–2025) promoting artificial intelligence (AI) in digital cultural industries (DCIs). Focusing on policy synergy for sustainable AI, the analysis assesses ten primary variables—including policy objectives, instruments, sustainability integration, and application levels—across national and local initiatives. Results reveal an overall ‘Excellent’ average PMC score, indicating robust policy design in economic or technological domains. However, some low-grade policies prioritize short-term regulation over sustainable governance. The study recommends embedding ‘human-centricity’ and ‘ethical risk assessment’ in sustainability indicators, alongside cultural diversity safeguards (e.g., algorithmic support for intangible heritage). These optimizations can reframe AI as a socio-cultural enabler—not merely a technical tool—advancing equitable, innovative, and ecologically resilient DCIs aligned with Sustainable Development Goals (SDGs).
Citation: Qu C, Zhao X (2026) Policy Modeling Consistency index-based study on policy synergy for sustainable artificial intelligence in China’s digital cultural industries. PLoS One 21(3): e0345004. https://doi.org/10.1371/journal.pone.0345004
Editor: Pengfei Fu, Shanghai Jiao Tong University, CHINA
Received: June 5, 2025; Accepted: February 28, 2026; Published: March 16, 2026
Copyright: © 2026 Qu, Zhao. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results presented in the study are available from GitHub at: https://github.com/ChenQu-ola/-PMC-index-based-study-on-policy-synergy-for-sustainable-ai-in-China-s-digital-cultural-industries.
Funding: This research was funded by Major Program of the National Social Science Foundation of China [23&ZD087].
Competing interests: The authors have declared that no competing interests exist.
Introduction
Artificial intelligence (AI) is rapidly reshaping the global socio-economic landscape. Beyond enhancing efficiency through algorithms and computing power, AI is crucially a ‘socio-cultural technology’ [1], redefining how human knowledge and innovation are shared, transmitted and recreated. Within the framework of the Sustainable Development Goals (SDGs), the application of AI should transcend mere technical utility and focus on creating social value that promotes equity, innovation, and cultural inclusivity [2]. This process highlights the critical role of policy guidance, particularly in the integration of digital technology and creative industries. The deep interface of AI with cultural heritage, artistic creation, and media dissemination offers new pathways for the preservation, transmission, and recreation of collective human knowledge [3,4].
This paper evaluates and optimizes policies which highlights the role of sustainable AI within the context of China’s digital cultural industries (数字文化产业). According to China’s official document, the DCIs focus on cultural and creative content which rely on digital technology for creation, production, dissemination and services [5]. China is experiencing its digital transformation of cultural industries [6]. Particularly, China has recently implemented a series of innovative policies in the sector of DCIs to embed AI technology deeply within the cultural production and dissemination chain [7]. This includes using virtual reality (VR) to revitalize historical artifacts, employing generative AI to democratize content creation, and leveraging large models for precise international distribution of cultural products [8].
However, AI-driven cultural innovation also faces significant challenges, such as data privacy risks, algorithmic biases leading to information bubbles, and copyright disputes over AI-generated content [9]. These issues require systematic policy responses. In this context, this study utilizes the Policy Modeling Consistency (PMC) [10] index model to quantitatively evaluate China’s DCIs policies, aiming to reveal strengths and weaknesses in policy design and explore how to optimize policies to foster the synergistic development of AI technology and the cultural industries, ultimately achieving multidimensional sustainable development goals.
Beyond its empirical evaluation, this study offers a conceptual contribution to emerging debates on sustainable AI and socio-cultural technologies [1,11]. By applying the PMC model to policies governing AI in digital cultural industries, we demonstrate that AI in China is framed not only as a technological instrument but as a cultural and sustainability infrastructure that shapes how knowledge, values, and social practices are reproduced. This perspective extends existing discussions of sustainable AI, which often emphasize environmental efficiency or ethical risk, by foregrounding the cultural dimensions of AI governance and highlighting how policy design mediates the relationship between technological innovation, cultural vitality, and long-term socio-ecological goals. The PMC analysis reveals structural imbalances in policy orientation, showing that economic and technological objectives consistently outweigh cultural and sustainability considerations. This finding enables us to propose a normative framework that integrates human-centric governance, cultural diversity protection, and ethical risk assessment into the design of sustainable AI policies. Through this theoretical lens, the paper advances the understanding of AI as a socio-cultural technology and illustrates how policy systems can more effectively align AI innovation with cultural sustainability and the SDGs.
Literature review
Sustainable artificial intelligence: Concepts, opportunities, and risks
AI has been widely recognized as the driver of sustainable development, with scholars arguing that AI can enhance efficiency, reduce environmental impact, and address complex global challenges [2,12]. From applications in energy management to healthcare systems, AI contributes to multiple Sustainable Development Goals (SDGs), offering opportunities to optimize resource use, improve decision-making, and bridge global development gaps [13,14]. The World Economic Forum has mapped hundreds of technology applications to SDGs, underscoring AI’s relevance across environmental, economic, and social dimensions [15]. However, AI’s rapid integration into society also generates new risks: resource-intensive computation, algorithmic opacity, social bias, and inequalities stemming from uneven technological access [16]. These challenges illustrate that AI’s sustainability contributions cannot be understood as purely technical achievements but account for broader socio-economic impacts.
To address these tensions, some researchers have advanced the framework of Sustainable Artificial Intelligence (SAI), which considers AI both as a tool for achieving SDGs and as a system requiring governance to ensure its own sustainable use [11,17]. SAI research emphasizes human-centricity, accountability, environmental responsibility, and ethical design, highlighting the need for multi-level governance mechanisms to prevent undesirable societal outcomes [18]. Farrell, Gopnik (1) argue that AI should be analyzed not merely as an intelligent agent but as a socio-cultural technology, embedding human knowledge, norms, and institutional logics. This perspective suggests that successful SAI implementation depends on organizational culture, regulatory structures, and societal expectations—not only on technological capability [19].
Cultural sustainability and AI in the digital age
The integration of digital technologies into cultural sectors has transformed the production, dissemination, and preservation of cultural knowledge. In China, the concept of cultural industries is derived from creative industries, and sometimes the two are used interchangeably [20]. As a later coined term, digital cultural industries (DCIs) emerged as a response to the growing role of digital tools in cultural innovation, heritage protection, and creative content production [21]. Similar global initiatives—including the UK’s CreaTech [22], Hong Kong’s Arts+Tech strategy [23], and UNCTAD’s Creative Industries 4.0 [24]—likewise position digital technologies as engines of cultural sustainability. These frameworks collectively suggest that AI and related technologies (e.g., VR, AR, generative models) play a pivotal role in reimagining cultural participation, enabling immersive heritage experiences, and expanding creative capacities [3,4].
However, the cultural implications of AI also raise concerns. Biases embedded in datasets risk reinforcing dominant cultural narratives, potentially marginalizing dialects, minority cultures, or intangible heritage [7,25]. Algorithmic personalization may narrow cultural diversity by prioritizing commercially successful or mainstream content. Moreover, intellectual property and authenticity debates surrounding AI-generated cultural content complicate long-term sustainability [9]. Scholars have therefore called for culturally sensitive AI governance frameworks that recognize both the creative possibilities and socio-cultural risks of digital transformation.
In China, these tensions are particularly pronounced due to strong state interest in leveraging AI for soft power, digital innovation, and industrial upgrading [26,27]. National strategies increasingly emphasize integrating AI across the cultural value chain—from heritage digitization to intelligent content distribution to VR-enhanced cultural industries [28]. At the same time, cultural institutions are experimenting with AI-generated cultural products, raising questions about cultural authenticity, community participation, and long-term sustainability [29].
The discussion on cultural sustainability in the era of AI highlights that AI’s value in cultural industries cannot be evaluated solely through economic or technological metrics. Instead, policies must be assessed for their support of cultural diversity, heritage preservation, equitable participation, and intergenerational transmission of cultural knowledge. These considerations directly inform the selection and interpretation of the policy evaluation related to sustainability and policy functions.
Policy evaluation and the PMC framework
This section concerns methods for evaluating public policies, particularly in complex and multi-dimensional domains such as AI governance and cultural industries. Traditional approaches—including logic models, regulatory impact assessments, and qualitative content analysis—offer valuable insights but are often limited in their ability to capture multidimensional policy interactions [30,31]. The Policy Modeling Consistency (PMC) index, developed by Ruiz Estrada (10), addresses this gap by providing a systematic method for quantifying internal coherence across multiple policy components. The PMC model has been applied in diverse areas, including renewable energy, biotechnology, education, talent management, and social welfare [32,33]. Its flexibility makes it particularly suitable for analyzing policy synergy and internal structure, enabling researchers to compare policies across regions, time periods, or thematic domains.
Within cultural industry research, the PMC model has recently been adapted to assess policy design and coherence in China [34]. This emerging methodological trend demonstrates its growing relevance to cultural policy studies. Importantly, PMC’s structure—built around evaluating objectives, instruments, sustainability dimensions, and functions—aligns well with the theoretical pillars outlined above. It provides a way to assess whether policies articulate sustainability principles (as emphasized in SAI), incorporate cultural sustainability considerations, and deploy coherent policy tools to achieve desired outcomes.
Thus, the PMC index constitutes provides both the methodological rationale and the analytical lens through which China’s AI-related DCIs policies are evaluated. By bridging the conceptual insights from SAI and cultural sustainability with a quantitative evaluation model, this study integrates theoretical rigor with empirical precision.
Methods
This research focuses on the quantitative research of policies related to China’s DCIs, employing the PMC index model to analyze policies from 2017 to 2024. It aims to construct targeted models, evaluation standards, calculation methods, and grading assessments for sustainable AI policies within the DCIs, proposing quantitative evaluation and optimization methods. The PMC index model was first introduced by Mario Arturo Ruiz Estrada [10]. This model helps researchers and decision-makers determine the consistency level of policies and assess the strengths and weaknesses of the policies to be evaluated within the policy index model. The PMC has been widely applied in sustainable research fields such as pharmaceutical policies, renewable energy environmental policies, educational policies, talent policies, and social policies [32,33]. Recently, it has also begun to be used in studies of cultural industries policies in China [34]; however, these studies have not addressed the impact of digital technology on the transformation of China’s cultural industries, and there is very little research specifically on sustainable AI in the context of China’s DCIs.
Construction of the PMC index model
This paper defines the sample time for digital cultural industries policies as 2016–2025. The 2016 China Government Work Report [35] saw the central government propose the concept of ‘digital creative industries’, which was later amended in April 2017 to ‘digital cultural industries’. Furthermore, this paper outlines China’s DCIs policies, selecting those highlighting AI. This selection is based on the innovative, experimental, and demonstrative nature of AI policies during this period, which is conducive to constructing the PMC index model for quantitative evaluation and optimization. The primary sources for policy collection include the official website of the Central People’s Government, the National Bureau of Statistics, and local government portals, and supplemented by search engines like Peking University Law Database (北大法宝).
To ensure transparency and consistency in policy selection, explicit inclusion and exclusion criteria were applied [36,37]. A policy document was included if it (1) was issued by a national or provincial government authority, (2) explicitly referenced artificial intelligence or AI-enabled digital cultural technologies, (3) contained concrete policy measures suitable for PMC indicator coding, and (4) fell within the domain of cultural industries or cultural governance. Documents were excluded if they mentioned AI only superficially, lacked codable content, were duplicate announcements, or represented non-policy materials such as press releases or meeting summaries. Following these criteria, an initial pool of 82 documents underwent title-level screening, abstract-level relevance checks, and full-text assessment. 32 policies met all eligibility criteria and were included in the final analysis (see Table 1). To enhance coding reliability, a second reviewer independently assessed a sample of the policies using the predefined coding rubric, and discrepancies were resolved through discussion. A flow diagram of the selection process has been added to provide a clear overview of the document filtering procedure (Fig 1).
Following the variable settings proposed by Estrada, we selected ten primary variables: (X1) Policy Nature, (X2) Policy Timeframe, (X3) Policy Domain, (X4) Policy Objectives, (X5) Policy Instruments, (X6) Level of Application, (X7) Incentive Measures, (X8) Sustainable Development, (X9) Policy Functions, and (X10) Citation Recourse. Moreover, to ensure the maximal completeness and accuracy of statistical analysis, word segmentation was performed on 82 DCIs policies from 2016 to 2025 using the ‘Jieba’ library in Python, resulting in a list of the top 90 high-frequency terms including top 30 two-character Chinese words, to 30 three-character Chinese words and 30 four-character Chinese words (see Table 2). Based on high-frequency word statistics, secondary variables under X4 to X9 were established. The secondary variables for X1 to X3 and X10 are based on Estrada’s original model index and relevant PMC studies. Finally, this paper formed a PMC index model for evaluating digital cultural industries policies from 2016 to 2024, comprising 10 primary variables and 35 secondary variables (see Table 3).
Variables X1 to X10 represent nine primary dimensions for evaluating the role of AI in DCIs policies, forming a comprehensive assessment framework these policies.
- X1 (Policy Nature) distinguishes between normative, guiding, strategic, and pilot policies, reflecting the fundamental positioning and function of each policy.
- X2 (Policy Timeframe) covers short-term, medium-term, and long-term policies, helping to assess implementation cycles and expected impacts.
- X3 (Policy Domain) identifies the main focus areas of the policy, such as culture, technology, economy, or society.
- X4 (Policy Objectives) clarify the intended goals, including technological innovation, expanding consumption, talent development, and regulation introduction, highlighting the policy’s key priorities.
- X5 (Policy Instruments) includes financial support, regulatory standards, public services, and market cultivation, providing practical instruments for policy implementation.
- X6 (Level of Application) refers to the level at which the policy is formulated and applied—national, local, or industry level—indicating the scope and extent of its influence.
- X7 (Incentive Measures) focuses on incentives related to talent, capital, and technology, which are crucial for enhancing the attractiveness and effectiveness of the policy.
- X8 (Sustainable Development) emphasizes green development, long-term development, and high-quality growth, reflecting the strategic foresight embedded in the policy.
- X9 (Policy Function) examines areas such as industrial integration, technological empowerment, regional cooperation, and collaborative innovation, showcasing the policy’s role in transforming and upgrading the cultural industry.
- X10 (Citation Recourse) evaluates whether relevant literature is cited within the policy to support its rationale and legitimacy.
Taken together, these variables form a systematic and structured framework for analyzing and categorizing DCIs policies.
PMC index calculation and measuring method
The data in this paper is based on the content of policies, with corresponding values assigned to lower-level variables to derive the relevant data. This research uses Python scripts to assign values to each secondary variable and analyzes whether the high-frequency terms from the selected 32 DCIs policies align with the variable setting standards.
According to the principles established for the PMC index system, the following rules are defined: 1 for meeting the standard and 0 for not meeting the standard. For example, in the case of indicator X8 (Sustainable Development), each secondary indicator was designed based on high-frequency terms identified through policy text mining, while the specific scoring keywords were selected after reviewing all 32 policy documents and taking into account Chinese linguistic semantics and policy discourse conventions. Under X8, the secondary indicator X8:1 “Green Development” was operationalized using keywords such as “green” (绿色), “low-carbon” (低碳), and “environmental protection” (环保). The secondary indicator X8:2 “Long-term Development” was coded based on terms including “sustainability” (可持续), “long-term mechanism” (长效机制), and “future-oriented” (未来). The secondary indicator X8:3 “High-quality Development” was identified through keywords such as “high-quality” (高质量), “modernization” (现代化), “institutional soundness” (健全), and “diversification” (多元化). When one or more of these predefined keywords appeared in a policy document in a context consistent with the corresponding secondary indicator, the indicator was assigned a value of 1; otherwise, it was coded as 0. This rule-based and programmatic approach ensured consistency and reduced subjective interpretation in the scoring process.
Following the methodology proposed by Ruiz Estrada, the calculation model involves four steps: First, secondary variables are positioned as multiple inputs and outputs, and empirical analysis is used to assign values to these variables. Second, all secondary variables are constrained to follow a distribution within the [0,1] interval, with the assignment methods detailed in formulas (1) and (2). Third, as shown in formula (3), each first-level variable is calculated as the average sum of its corresponding secondary variables. Finally, formula (4) is applied to compute the PMC index for each indicator.
t=一级变量 j=二级变量
Based on the above Formulas (1)-(4), this study calculated the PMC index scores of cultural industry policies. These scores were then used to classify all evaluated policies into graded categories. Since the evaluation framework incorporates 10 first-level variables, the PMC index ranges from 0 to 10. The classification criteria follow Estrada’s grading standard for cultural policy PMC index scores (Table 4).
To scientifically and comprehensively evaluate the DCIs policies, this research employs the PMC surface construction method based on the mentioned evaluation criteria and index scores for comparative analysis. The PMC surface visualizes all results contained in the PMC matrix, intuitively revealing the strengths and weaknesses of the policies under review within a multidimensional coordinate space, thereby more vividly reflecting the overall policy performance. The PMC surface is constructed based on a 3 × 3 variable PMC matrix, which encompasses independent scores for nine key variables. According to the classification of DCIs policy grades, the closer the PMC surface approaches the upper value of 1, the better the PMC index score of the policy. When the scores of the 3 × 3 variable matrix tend to be balanced and approach 1, the PMC surface displays an ideal symmetrical shape.
This paper defines a total of ten first-level variables, among which variable X10 represents the policy document reference value. Since none of the sampled policies mention reference documents, this variable scores 0 across all policies, lacking differentiation. The X10 (Citation Recourse) variable was removed from the PMC matrix because all sampled policy documents scored zero on this indicator. This uniformity reflects not a deficiency in the policies themselves, but a structural characteristic of Chinese policy writing [38], which rarely employs explicit citations or reference systems commonly found in Western governance documents. The absence of variance means that X10 does not contribute analytically to distinguishing policy performance, and retaining such a variable would distort the overall PMC evaluation. Therefore, variable X10 is excluded, and the PMC matrix is constructed using the scores of the remaining nine first-level variables, as shown in Formula (5). This approach effectively eliminates the interference of the policy document reference value on the overall score, allowing for a more accurate reflection of the intrinsic quality and structural characteristics of the policies.
Results: policy text rating and PMC surface drawing
Policy text rating
Based on Formulas (1), (2), (3), and (4), this research calculates the PMC index scores for DCIs policies. Additionally, a Pavg value (average PMC score across 32 policies) was included. Following the grading criteria in Table 5, policies enacted between 2016 and 2020 were classified into respective grades.
From the 32 policy samples, four representative policies (P12, P10, P19, P1) – selected as one per grade (A, B, C, D) – along with the Pavg value were analyzed in PMC surface plots. The five PMC surface plots (Figs 2–6) were generated to comparatively examine variable-dimensional performance across policies.
To explore whether systematic differences exist in policy quality, additional subgroup analyses were conducted across issuing authority, region, and policy release period. Specifically, average PMC index scores were compared between policies issued by central and local governments, between earlier (2016–2022) and more recent (2023–2025) policy cohorts, and across major regional groupings (see Table 6). The regional groupings follow the conventional macro-regional classification in China by categorizing coastal regions with a high degree of openness and more mature economic and industrial foundations as eastern regions (e.g., Shanghai, Jiangsu, Zhejiang, and Guangdong), while classifying predominantly inland regions at a relatively earlier stage of development and primarily responsible for industrial transfer and regional coordination as central and western regions (e.g., Xi’an, Chengdu, and Lv’an).
Overall, the results suggest that differences in average PMC scores across these dimensions are moderate rather than pronounced. Local government policies exhibit slightly higher average PMC scores than central government policies, and policies released in the later period show a modest improvement compared with earlier ones. Regional comparisons likewise reveal relatively close score distributions between eastern and central–western regions. Taken together, these patterns indicate that variations in policy quality are not driven by a single structural attribute, but instead reflect a broadly convergent policy design logic across issuing levels, regions, and time periods.
Discussion: quantitative analysis of policy texts related to artificial intelligence in the digital cultural industry
The PMC-based evaluation of 32 policies governing AI in China’s DCIs reveals important structural patterns that carry significant policy implications. According to the analyses across issuing authority, region, and policy release period, the PMC results reveal not variation in policy quality, but a structural tendency toward techno-economic prioritization. The challenge is not uneven policy quality, but a shared structural blind spot.
Although the overall policy landscape demonstrates some strength, particularly in its emphasis on technological innovation, industrial development, and economic growth, the analysis also identifies persistent weaknesses in sustainability-related indicators. These gaps suggest that existing policies often conceptualize AI primarily as an engine of economic and technical modernization, rather than as a socio-cultural technology requiring long-term, human-centered governance [1,18]. As the results show, the sustainable development dimension (X8) scores consistently lower than other variables, indicating that environmental considerations, ethical safeguards, and cultural sustainability are not yet systematically embedded across policy instruments. This imbalance highlights a need to reorient AI governance frameworks so that they align more closely with the principles of SAI articulated in current scholarship [16,17].
The comparison of representative policies across A, B, C, and D grades illustrates these challenges more concretely. High-performing policies such as Shanghai’s Metaverse Action Plan (P12) embed sustainability by integrating green data centers, long-term development mechanisms, and coordinated fiscal tools, demonstrating that coherent policy design is achievable when sustainability is understood as foundational rather than supplementary. By contrast, lower-ranked policies, especially those falling into the C and D categories, relay heavily on regulatory controls or short-term incentives without establishing mechanisms for ethical concerns, cultural protection, or intergenerational equity. Such policies treat AI as an isolated technological intervention rather than as part of a broader socio-cultural system, an approach that risks undermining cultural diversity, exacerbating algorithmic bias, and narrowing the social benefits of AI [9,25].
These results point to several actionable suggestions for policymakers. Sustainability indicators should be expanded and standardized across national and local policy instruments. Current frameworks pay limited attention to cultural sustainability, despite clear evidence that AI reshapes aesthetics, creative practices, and heritage systems [7,39]. Policies could explicitly incorporate cultural diversity metrics, such as algorithmic support for dialect processing, minority cultural representation in datasets, or incentives for AI-driven intangible heritage revitalization. Embedding such indicators would ensure that DCIs grow in a manner that supports diversity and inclusivity.
Furthermore, the findings indicate that policy instruments need to be rebalanced to support long-term socio-cultural outcomes. While financial incentives and industrial development measures are widely deployed, they should be complemented with public service tools, community participation mechanisms, and transparent oversight structures. These instruments would enable policymakers to anticipate unintended consequences, strengthen accountability, and facilitate collaborative governance across cultural institutions and technology firms. Such an approach echoes broader recommendations in international SAI scholarship, which emphasize the social, behavioural, and institutional determinants of sustainable AI deployment [12,19].
Finally, although the present study focuses on China, the implications extend to global debates on AI governance. Many countries face challenges similar to those identified here: rapid technological adoption outpacing ethical regulation, uneven integration of sustainability principles, and growing tensions around cultural representation in AI systems. The PMC-based framework developed in this study offers a transferable method for evaluating policy coherence across diverse governance contexts. By quantifying gaps in sustainability, cultural protection, and long-term planning, the model can support international policymakers in designing approaches that balance innovation with social responsibility. Aligning national AI policies with global standards, such as the United Nations’ SDGs [40] and human rights–based AI frameworks [11] would further promote policy convergence and collective learning across regions.
Conclusion
This study demonstrates that China’s AI-related policies for digital cultural industries possess a solid foundation in supporting technological advancement and economic modernization, yet reveal significant gaps in sustainability integration. While the high PMC scores in objectives, instruments, and technological empowerment indicate strong institutional momentum, the persistently weak performance in sustainability dimensions points to an underdeveloped awareness of AI’s broader socio-cultural and ethical implications. The evaluation of representative policies confirms that the most effective governance frameworks are those that embed sustainability principles across multiple policy dimensions linking long-term planning, cultural diversity, ethical oversight, and cross-sector collaboration. Such integration transforms AI from a productivity tool into a genuinely socio-cultural technology capable of supporting the multidimensional aims of the SDGs.
The insights generated by this study carry broader relevance beyond China. Many countries are grappling with similar tensions between rapid digital transformation and the need for responsible, inclusive governance. The PMC index, by offering a structured means of diagnosing inconsistencies and identifying areas for improvement, provides a valuable evaluative framework for policymakers worldwide. Its multidimensional approach makes it particularly well suited for sectors, such as cultural industries where technological, social, ethical, and cultural factors intersect. By adopting evaluation tools that highlight not only technological readiness but also cultural and sustainability considerations, policymakers can better shape AI development toward equitable and culturally grounded futures.
In moving forward, China’s DCIs policies, and international AI governance more broadly, would benefit from incorporating explicit sustainability metrics, algorithmic accountability mechanisms, and cultural diversity safeguards. Such measures align with emerging visions of AI as a socio-cultural technology [1] and support global efforts to create human-centered, ethically grounded digital ecosystems. Strengthening these dimensions will not only enhance policy coherence but also ensure that AI contributes meaningfully to cultural vitality, social well-being, and long-term sustainable development.
At the same time, this study has limitations. The PMC model evaluates policy coherence but cannot fully capture implementation outcomes, local enforcement dynamics, or stakeholder perspectives. The coding of sustainability and cultural indicators, though grounded in text-mined evidence, is necessarily interpretive, and future research would benefit from triangulating PMC results with interviews, case studies, or longitudinal impact assessments.
Overall, this research underscores the need for AI policies that integrate sustainability, cultural inclusivity, and ethical governance as core principles rather than peripheral considerations. Future studies should build on the PMC approach to examine multi-level policy interactions, track changes across time, and contribute to the development of globally informed frameworks for sustainable and culturally responsible AI governance.
References
- 1. Farrell H, Gopnik A, Shalizi C, Evans J. Large AI models are cultural and social technologies. Science. 2025;387(6739):1153–6. pmid:40080578
- 2. Goralski MA, Tan TK. Artificial intelligence and sustainable development. Int J Management Education. 2020;18(1):100330.
- 3. Allal-Chérif O. Intelligent cathedrals: Using augmented reality, virtual reality, and artificial intelligence to provide an intense cultural, historical, and religious visitor experience. Technological Forecasting and Social Change. 2022;178:121604.
- 4. Mazzone M, Elgammal A. Art, Creativity, and the Potential of Artificial Intelligence. Arts. 2019;8(1):26.
- 5.
Ministry of Culture. 关于推动数字文化产业创新发展的指导意见(Guiding Opinions on Promoting the Innovation and Development of the Digital Cultural Industry). 2017. https://www.gov.cn/gongbao/content/2017/content_5230291.htm
- 6.
Hong Y. Networking China: The Digital Transformation of the Chinese Economy. Illinois: University of Illinois Press. 2017.
- 7. Liu Y, Song P. Creating Sustainable Cultural Industries: The Perspective of Artificial Intelligence and Global Value Chain. J Environ Public Health. 2022;2022(6768388):1–11.
- 8.
Zhao X. Digital immersive art in China: rejuvenation and cultural presence. Keane M, Haiqing Y, editors. London: Anthem Press. 2024.
- 9. Jorzik P, Klein SP, Kanbach DK, Kraus S. AI-driven business model innovation: a systematic review and research agenda. Journal of Business Research. 2024;182:114764.
- 10. Ruiz Estrada MA. Policy modeling: definition, classification and evaluation. J Policy Modeling. 2011;33(4):523–36.
- 11. Fernández-Aller C, Velasco AF, Á M, Pastor-Escuredo D, Pickin S, Criado JS, et al. An inclusive and sustainable artificial intelligence strategy for Europe based on human rights. IEEE Technol Society Magazine. 2021;40(1):46–54.
- 12. Nishant R, Kennedy M, Corbett J. Artificial intelligence for sustainability: Challenges, opportunities, and a research agenda. Int J Information Management. 2020;53:102104.
- 13. Leal Filho W, Yang P, Eustachio JHPP, Azul AM, Gellers JC, Gielczyk A. Deploying digitalisation and artificial intelligence in sustainable development research. Environment, Development and Sustainability. 2023;25(6):4957–88.
- 14. Liengpunsakul S. Artificial Intelligence and Sustainable Development in China. The Chinese Economy. 2021;54(4):235–48.
- 15.
OxfordInsights. Government AI Readiness Index 2024. https://oxfordinsights.com/ai-readiness/ai-readiness-index/
- 16.
Isensee C, Griese K-M, Teuteberg F. Sustainable artificial intelligence: A corporate culture perspective. Sustainability Management Forum | NachhaltigkeitsManagementForum. 2021;29(3):217–30. https://doi.org/10.1007/s00550-021-00524-6
- 17. Di Vaio A, Hassan R, Alavoine C. Data intelligence and analytics: A bibliometric analysis of human–Artificial intelligence in public sector decision-making effectiveness. Technological Forecasting and Social Change. 2022;174:121201.
- 18.
Mazzi F, Floridi L. The ethics of artificial intelligence for the sustainable development goals. New York: Springer. 2023.
- 19. Kulkov I, Kulkova J, Rohrbeck R, Menvielle L, Kaartemo V, Makkonen H. Artificial intelligence ‐ driven sustainable development: Examining organizational, technical, and processing approaches to achieving global goals. Sustainable Development. 2023;32(3):2253–67.
- 20. Keane M. Creative industries in China: four perspectives on social transformation. International Journal of Cultural Policy. 2009;15(4):431–43.
- 21. Xiaojuan J. Technology and Culture in the Digital Era. Social Sciences in China. 2023;44(1):4–24.
- 22.
CIC C. CreaTech is “huge national opportunity” to add value. https://www.thecreativeindustries.co.uk/site-content/createch-is-huge-national-opportunity-says-cic-co-chair. 2023.
- 23.
CCIDAHK. Integrated “Arts Tech”. https://www.ccidahk.gov.hk/en/art_tech.html. 2024.
- 24.
Escaith H. Creative industry 4.0: Towards a new globalized creative economy. 2022. https://unctad.org/publication/creative-industry-40-towards-new-globalized-creative-economy
- 25. Taurino G. Algorithmic art and cultural sustainability in the museum sector. Philosophical Studies Series. Springer International Publishing. 2023. p. 327–45.
- 26. Li F, Zhao X. Digital creative industries and the promotion of state cultural soft power: strategies and pathways. Journal of Guangxi University for Nationalities. 2017;39(6):2–7.
- 27. Zhao X, Keane M. Sino-futurism and alternative imaginaries of digital China. Media International Australia. 2023.
- 28. Wang M, Huang Y. A digital Technology–Cultural resource strategy to drive innovation in cultural industries: A dynamic analysis based on machine learning. Technology in Society. 2024;77:102590.
- 29. Zhang B, Cheng P, Deng L, Romainoor NH, Han J, Luo G, et al. Can AI-generated art stimulate the sustainability of intangible cultural heritage? A quantitative research on cultural and creative products of New Year Prints generated by AI. Heliyon. 2023;9(10):e20477. pmid:37810838
- 30. Riley BL, Kernoghan A, Stockton L, Montague S, Yessis J, Willis CD. Using contribution analysis to evaluate the impacts of research on policy: Getting to ‘good enough’. Research Evaluation. 2017;27(1):16–27.
- 31. Mills T, Lawton R, Sheard L. Advancing complexity science in healthcare research: the logic of logic models. BMC Med Res Methodol. 2019;19(1):55. pmid:30871474
- 32. Qi L, Chen W, Li C, Song X, Ge L. Quantitative Evaluation of China’s Biogenetic Resources Conservation Policies Based on the Policy Modeling Consistency Index Model. Sustainability. 2024;16(12):5158.
- 33. Zhang C, Li X, Sun Y, Chen J, Streimikiene D. Policy modeling consistency analysis during energy crises: Evidence from China’s coal power policy. Technological Forecasting and Social Change. 2023;197:122931.
- 34. Hu P, Zhang Z. Quantitative evaluation and optimisation of cultural industry policies based on PMC index model. Journal of Arts Management. 2024;(01):66–77.
- 35.
Li K. Report on Government Work (2016年中国政府报告). State Council of China. 2016. http://www.gov.cn/guowuyuan/2016-03/17/content_5054901.htm
- 36. Bowen GA. Document analysis as a qualitative research method. Qualitative Res J. 2009;9(2):27–40.
- 37.
Dunn WN. Public Policy Analysis. New York: Routledge. 2015.
- 38. He B, Warren ME. Authoritarian deliberation: the deliberative turn in chinese political development. Perspect polit. 2011;9(2):269–89.
- 39. Zhang Z, Pan Y. Research on visitors’ motivation and viewing effect of immersive exhibition from the perspective of ERG theory. Cultural Industry Research. 2023;1:2–20.
- 40.
UnitedNations. Sustainable development goals. 2020. Available from: https://sdgs.un.org/goals