Artificial Intelligence and Journalistic Practices in China: A Grounded Theory Analysis Based on Qualitative Interviews ()
1. Introduction
1.1. Background
The advent of Artificial Intelligence (AI) has changed journalism worldwide and how news is made, spread, and received. Artificial intelligence tools have been embraced by journalists to perform tasks such as automated content creation, data processing, and audience interaction. For instance, these tools are used by big news outlets such as The New York Times and The Washington Post to enhance efficiency in routine reports like financial news, weather reports, and sports news (Jamil, 2021). Moreover, generative AI models such as ChatGPT and DALL-E are revolutionizing multimedia narratives by integrating textual and visual content to develop fresh and creative stories (Wang et al., 2023).
In the Chinese context, the adoption of AI in journalism is driven by the country’s emphasis on technological advancement and state-led media regulation. AI technologies enable journalists to streamline content production while aligning with regulatory frameworks that prioritize state narratives (Ji et al., 2024). This dual emphasis highlights a unique intersection of innovation and control, where AI serves as both a tool for enhancing efficiency and a mechanism for enforcing political objectives (Sun et al., 2024).
AI also helps in the emergence of new positions in Chinese journalism. For instance, AI news anchors are being used more frequently, which is an advancement in real-time audio-visual synthesis, but it also brings in questions of authenticity and transparency (Xue et al., 2022). These developments show how AI transforms journalistic activities and how it impacts audience trust in the media.
AI in journalism has implications that go beyond efficiency in the global context. Scholars maintain that AI adoption is an expression of sociotechnical imaginaries that create both benefits and drawbacks across different cultures and political systems (Barredo Ibáñez, Jamil, & de la Garza Montemayor, 2023). In China, these imaginaries are linked to the state’s employment of AI for content filtering and the propagation of state-approved information, which illustrates the dual character of AI as an innovation facilitator and a monitoring device (Wang et al., 2023).
This study is especially important because it explores how AI tools are incorporated into Chinese newsrooms to identify the patterns, problems, and prospects of this specific media environment. Thus, the study aims to contribute to the understanding of the potential of AI in journalism transformation.
1.2. Research Problem and Questions
The adoption of AI in journalism has several important questions, particularly in a place like China, where technology meets politics and culture. This paper aims to answer the following main questions:
1) How does AI change the work of journalists in China?
2) What ethical, cultural, and regulatory issues are related to AI in journalism in China?
3) How do journalists view and welcome the growing use of AI in their profession?
The research aims to enhance our understanding of AI effects on journalism while exploring its transformative nature and diverse implications across different media settings.
1.3. Objectives
The main goal of this research requires Grounded Theory application to reveal basic patterns and frameworks. This study examines qualitative data to deliver a detailed analysis of how AI transforms journalistic work while focusing on the Chinese framework of regulatory and cultural influences. The research results will help both scholars and practitioners address the challenge of maintaining technological progress alongside ethical standards and professional accountability.
2. Literature Review
2.1. Artificial Intelligence in Journalism
The media industry undergoes transformation because of artificial intelligence (AI). AI technologies, particularly Natural Language Generation (NLG), automate news production from datasets, thereby enabling content creators and journalists to bypass repetitive and time-consuming tasks. Major news organizations BBC, The Washington Post, and The New York Times utilize AI tools for basic news production to enhance operational effectiveness and resource management (Jamil, 2021; Gong, 2023) while human journalists maintain their essential role. The same time, ChatGPT and DALL-E expand journalism’s creative possibilities for storytelling (Wang et al., 2023).
The implementation of AI in newsrooms has transformed the way journalistic workflows function. Journalists can perform investigative work and creative tasks because automated tools handle routine duties (Kuai, 2024). News organizations such as The Guardian and Forbes employ AI systems to provide accurate live coverage of stock market fluctuations. The new technology prompts fundamental questions about both the genuineness and responsibility of materials generated through AI (Jia, Riedl, & Woolley, 2024).
AI algorithms have become essential for delivering personalized news consumption experiences to users. The algorithms process user information to select content that matches individual tastes, which leads to better user engagement (Sun et al., 2024). These systems help produce filter bubbles alongside information echo chambers that strengthen personal biases while reducing global perspectives (Ji et al., 2024). The ethical challenges of content selection create obstacles for standard journalistic practices that emphasize impartial news coverage.
Interactive AI tools such as chatbots and visualizations improve audience comprehension through simplified complex data-driven stories according to Kuai (2024). AI systems that generate visual content help make climate change statistics accessible for public engagement. The progress made in these areas generates new ethical problems regarding information disclosure and possible control methods (Jamil, 2023).
The implementation of AI technology boosts operational efficiency together with innovation but it comes with hazards such as algorithmic discrimination along with data integrity problems and copyright disputes. The utilization of proprietary algorithms in news production reduces journalistic oversight which creates difficulties for holding parties accountable (Jia et al., 2024). AI-generated news anchors that appear in China’s media showcase technological excellence as well as societal worries regarding human-like simulations potentially taking over authentic human communication (Xue et al., 2022).
AI journalism practices show different sociotechnical visions across worldwide regions. The Chinese government uses AI to control journalistic content which creates concerns about media independence (Ji et al., 2024). The Western media sector views AI as a tool to strengthen democratic discourse yet faces difficulties regarding disinformation while also addressing these issues (Barredo Ibáñez, Jamil, & de la Garza Montemayor, 2023). The different cultural approaches toward AI show why experts need to develop a refined understanding of AI’s effects within specific political settings.
The application of AI technology transforms journalism while acting as a disruptor within the industry. The new technology requires media professionals to reassess their work functions by merging technical advantage with moral responsibility. Research should investigate how artificial intelligence will affect society through its ability to improve journalistic practices and protect democratic principles and journalistic independence. The development of regulatory systems which guarantee transparency along with accountability stands essential to build trust in AI-driven journalistic practices.
2.2. Journalism in China: Challenges and Opportunities
The development of journalism in China is determined by specific set of factors: political, cultural and technological ones, which reflect the tension between the state control, cultural specifics and technological development. The government of China has significant influence on the media industry, making sure that journalism is consistent with the government agenda and social order. This regulatory environment restricts the freedom of the press and views the media as a means of conveying public information and supporting the realization of collective objectives (Jamil, 2023). Nevertheless, these limitations have led to the development of certain journalistic strategies, with journalists playing the role of both critics and supporters of the state discourse (Ji et al., 2024).
The rules that govern China tend to emphasize societal order and political tranquility, and therefore limit the freedom of the press. The operations of media houses are supervised and the message they communicate has to conform to the national agenda. This regulatory approach places the journalist in a dual position where they are expected to both criticize and endorse government policies. For instance, investigative journalism in China often highlights issues in the private sector, which in turn contributes to portraying the government as a defender of public interest (Ji et al., 2024). Such practices reveal how regulatory measures affect the extent and direction of journalistic work.
The Chinese media is characterized by cultural norms that emphasize the group and the society rather than the individual. This cultural frame channels journalists to tell stories that enhance the image of the nation and its growth agenda (Sun et al., 2024). Although this framework is useful for the propagation of unified national discourses, it may also hinder the freedom of the press especially when it comes to dealing with matters of contention or sensitive politics. Studies have shown that journalists practice self-censorship in order to avoid offending cultural and political sensibilities and to remain relevant in the society (Barredo Ibáñez, Jamil, & de la Garza Montemayor, 2023).
Technological advancements especially the use of Artificial Intelligence (AI) have been changing the landscape of Chinese journalism. AI tools boost the production of news through increasing the speed of news production and the ability to provide news content to different people (Kuai, 2024). For instance, AI-powered news anchors are slowly becoming a common feature of news presentation, which is a clear indication of technological advancement but also poses challenges in terms of realism and human-machine interaction (Xue et al., 2022). However, the dual purpose of AI as a means of innovation and as a means of surveillance also points to its multifaceted nature. This ambivalence is evident in the state’s use of AI for content analysis and story shaping, which is problematic given the lack of transparency in the use of algorithms and potential impact on journalistic ethics (Wang et al., 2023). Furthermore, while regulatory influences are widely discussed in the literature, other structural forces—such as commercial algorithms, market-driven content strategies, and internal newsroom cultures—also shape AI adoption. As Opdahl et al. (2023) argue, “technology is not simply an external tool journalists are forced to assimilate into newswork… but is instead a tool shaped by journalistic practices, needs and norms.”
The convergence of the regulatory, cultural, and technological factors results in a complex environment for journalism in China. Journalists are therefore required to work within the given rules while trying to explore the use of AI to create new content and connect with the target market. At the same time, they face challenges such as algorithmic bias, data privacy concerns, and limitations on press freedom (Jia, Riedl, & Woolley, 2024). These challenges require a more detailed comprehension of the ways in which innovation can be sustained without compromising ethical and professional standards.
The changing nature of Chinese journalism shows that it is important to develop strategies that are sensitive to the interplay of the regulatory, cultural, and technological factors. More research should be conducted on the impact of AI on journalism and the ways in which journalists can ensure their freedom and professionalism in the context of AI-generated content. In this way, policymakers, media organizations, and journalists will be able to better understand the changes taking place in the practice of journalism in China.
2.3. Gaps in Existing Research
Research on AI and journalism has expanded over time but researchers have not fully understood how AI transforms journalism in China. The literature review of AI adoption in newsrooms emphasizes its efficiency and automation capabilities but does not deeply examine how AI influences journalistic practices and professional ethics and audience perceptions in China’s environment.
Studies about AI technology and journalism functions do not investigate adequately how AI transforms Chinese journalists’ work roles and identities. The literature review shows that Chinese journalists serve as both watchdogs and facilitators of state objectives but it fails to study how AI affects these roles and generates new ethical challenges (Ji et al., 2024). The implementation of AI in investigative journalism produces conflicts with state narratives, which merit further investigation. (Jia, Riedl, & Woolley, 2024).
The majority of studies do not examine the sociotechnical imaginaries of AI in Chinese journalism. The public understanding of AI through narratives and perceptions about risks and benefits determines how AI implementation and governance take shape. Research about AI democracy promotion and disinformation fighting worldwide exists in abundance yet AI adoption in China needs thorough sociopolitical and sociocultural analysis (Barredo Ibáñez, Jamil, & de la Garza Montemayor, 2023).
This research area requires additional investigation because it lacks longitudinal and comparative studies that study the development of AI in Chinese journalism through time. Research studies about AI frequently analyze short cases and temporary events without exploring extensive patterns and prolonged effects (Kuai, 2024). The lack of Chinese journalism vs Global North journalism comparison studies creates a knowledge gap because these comparisons help researchers understand unique challenges and opportunities (Situmorang, 2024).
Research about AI-generated news and its reception by Chinese audiences remains underdeveloped because there is limited information available about how different population groups react to such content. Moreover, existing studies—and this paper included—tend to focus predominantly on the Chinese mainland, with limited comparative exploration of Hong Kong region, Macao region, or Taiwan region, where media systems and regulatory environments differ significantly. The findings from Western-based studies about young audience responses to emotional visual elements in AI-generated news need further adaptation to align with Chinese cultural and social frameworks (Zhang, Xu, & Ding, 2024).
Future research should prioritize qualitative methodologies, such as ethnographic studies and in-depth interviews, to capture the lived experiences of journalists and audiences in China. The combination of sociological and technological and cultural analysis will provide more comprehensive insights about how AI affects Chinese journalism and the wider media sector.
3. Methodology
3.1. Research Design
The main research methodology of this study uses Grounded Theory (GT) to analyze how Artificial Intelligence (AI) transforms news industry operations within China while affecting journalistic work practices. Grounded Theory stands as a research methodology that uses data collection to develop theory because it serves exploratory research well when theory development is minimal or when systematic theoretical frameworks are absent (Glaser & Strauss, 1967).
3.2. Participants
The participants (see Table 1) included journalism and communication practitioners of various ages and from diverse sectors in China, including journalists, editors, broadcasters, and program hosts. Participants also comprised content producers from TV stations, self-publishing media, radio stations, new media companies, and other organizations. In addition, the sample included Chinese university students who aspire to work in the journalism and communication industry in the future. Through in-depth interviews and other methods, we gained a comprehensive understanding of the current use of AI tools among Chinese news professionals and their perspectives from multiple dimensions.
Table 1. Situation of interviewees.
NAME |
GENDER |
AGE |
TIME (minutes) |
Interviewee 01 |
female |
21 |
13 |
Interviewee 02 |
female |
41 |
20 |
Interviewee 03 |
female |
48 |
22 |
Interviewee 04 |
male |
23 |
11 |
Interviewee 05 |
female |
21 |
14 |
Interviewee 06 |
male |
23 |
9 |
Interviewee 07 |
female |
21 |
17 |
Interviewee 08 |
female |
25 |
15 |
Interviewee 09 |
male |
23 |
10 |
Interviewee 10 |
female |
21 |
11 |
Interviewee 11 |
female |
21 |
19 |
3.3. Data Analysis
This study adopts the Grounded Theory approach to systematically analyze the interview data, employing Open Coding, Axial Coding, Selective Coding, and finally Theoretical Coding to construct the theoretical framework. This process ensures the extraction of core concepts from the data, revealing key patterns in AI applications in China’s news industry and their broader implications.
1) Open Coding: Identifying Initial Concepts
In the first step of data analysis, this research analyzed the interview texts line by line to identify and summarize the key phenomena and concepts mentioned by the interviewees. In this process, no theoretical framework is presupposed, but rather the data is broken down into smaller units of analysis by labelling key words, phrases and recurring themes. For example, when analyzing journalists’ perceptions of AI news production, concepts such as “increased efficiency”, “reduced human intervention”, and “content quality concerns” may emerge. These initial concepts provide the basis for subsequent categorization and theory building.
2) Axial Coding: Establishing Relationships between Concepts
Following open coding, this research categorized the initial concepts and explored the relationships between them. The core purpose of axial coding is to identify core categories and their attributes, and to establish causal, contextual, or interactional patterns between categories. For example, we might find a strong link between the categories “content automation” and “journalistic authenticity”, where the widespread use of AI has increased productivity but also raised questions about journalistic authenticity and ethics. This categorization allows us to understand more systematically how AI affects different aspects of journalism.
3) Selective Coding: Refining the Core Theory
In the Selective Coding stage, this research further integrated the results of the first two coding stages, selected core categories around the research theme, and constructed a theoretical framework. In this study, possible core concepts include “AI’s reshaping of the news production process,” “the changing professional identity of journalists,” and “the techno-ethical dilemma of journalism,” among others. The goal of this phase is to ensure that the research finds. The goal of this phase is to ensure that the findings are holistic and coherent and explain the broader impact of AI in journalism.
4) Theoretical Coding: Building a Theoretical Framework
Theoretical Coding is the final step in the Grounded Theory approach and aims to integrate all categories into a cohesive theoretical framework. By analyzing the deep interconnections among the data, we developed a theoretical model on the application of AI in China’s news industry. For instance, we may propose a “dual-role model of AI technology”; on the one hand, AI improves the efficiency of news production and facilitates the digital transformation of journalism; on the other hand, it introduces ethical challenges and leads to a redefinition of professional roles. This theoretical framework not only explains current phenomena but also provides guidance for future research.
3.4. Ethical Considerations
This study strictly follows the principles of ethical research to ensure the informed consent of the research subjects, the confidentiality of the data and the objectivity of the researcher. During the data collection process, all respondents voluntarily signed an informed consent form after fully understanding the purpose of the study, the research methodology and how the data would be used, to ensure the autonomy and transparency of their participation. In addition, in order to protect the privacy of the interviewees and data security, all interview data were anonymized, and coding methods were used in place of real identities during the research process to avoid leakage of personal information.
4. Findings
4.1. Open Coding, Axial Coding, and Selective Coding
We established the main phenomenon of the embedded growth of the AI technology in the Chinese news industry through Open Coding, Axial Coding, and Selective Coding. This development has not only changed the ways of news production but has also led to the changes in ethical, regulatory, and professional norms. Four major categories emerged from this core theme.
4.1.1. AI-Enabled Journalism: Efficiency Improvement and Innovation
AI has brought about remarkable improvements in efficiency and new approaches to news production, thus changing the course of journalism.
First, AI is critical in enhancing the production process of news. It accelerates the process of gathering, organizing, and processing information, thus making it easier for journalists and editors to work efficiently (Interviewee 11). Generative AI makes it possible to automate news writing, generate headlines, and summarize news, which helps reduce the monotony of work in newsrooms (Interviewees 09, 10).
Second, AI ensures the diversification of the news content. AI multimodal generation helps news move from text only to include images, videos, and audio to make it more engaging and easily consumable (Interviewee 06). Moreover, AI-based personalized news aggregation systems can offer news content that is relevant to the interests of the users, thus increasing user interaction and satisfaction (Interviewees 07, 08).
Last but not least, AI has brought about new forms of news delivery. AI-powered virtual news anchors and automated news production are being used in short video and social media platforms to provide the news industry with new ways of engaging the audience (Interviewee 05). Furthermore, Interactive Journalism (IJ) and Conversational Journalism (CJ) are being researched and developed to enhance the intelligence and personalization of news delivery (Interviewee 07).
The use of AI has cut the cost of making news, expedited the process of information delivery, and made the news industry more productive and diverse. However, this has also led to the erosion of authenticity and independence of the news content, which is a new challenge to the field of journalism.
4.1.2. Ethical Challenges and Limitations of AI News
Although AI shows enormous potential in enhancing news production efficiency, it still faces many ethical and limitation issues, mainly in the authenticity of news content, algorithmic bias and transparency, and professional ethical challenges brought by human-machine collaboration.
First, there have been numerous concerns raised about the credibility of the news produced by AI. AI may produce false or misleading information that can confuse people and damage the reputation of the media (Interviewee 08). AI’s language models are also prone to hallucination, producing content that seems plausible but contains factual errors (Interviewee 10). This concern aligns with Opdahl et al. (2023), who highlight that hallucinations arise from weak contextual modeling in large language models, making outputs appear convincing while lacking factual grounding. They also stress the need for provenance tracking—a system that allows journalists and readers to trace how information is generated and transformed—to address trust and verification challenges. Additionally, Amponsah and Atianashie (2024) warn that algorithmic biases may lead to echo chambers in AI-generated content, narrowing audiences’ exposure to diverse perspectives. These insights reinforce participants’ views that while AI tools are helpful, human oversight and transparency mechanisms are indispensable.
In addition, despite the fact that AI can be used to produce news automatically, fact-checking is still a human task, and AI cannot replace the professional skills of journalists (Interviewee 11). This means that although AI can increase productivity, the issue of the credibility and truthfulness of the information generated by AI remains a problem which requires human intervention to rectify.
Second, serious problems with the algorithmic bias and transparency threaten the impartiality of news. AI inherits cultural, political, or gender biases from the sources of its training data and, therefore, can compromise the objectivity of reporting (Interviewees 06, 08). Furthermore, recommendation algorithms can create a “filter bubble,” as users are only exposed to content that they like and thus do not have access to the other side of the argument (Interviewee 07). Most importantly, AI news is not always transparent, and it is therefore hard to tell where the information is coming from or whether it has been approved by some other credible entity (Interviewee 07). These problems can worsen the public’s distrust of media institutions.
Finally, the human-AI collaboration raises new challenges to the journalistic ethics. AI-assisted writing has become more popular, and this has changed the work of a journalist. The use of automated functions by AI may decrease the originality of reporting and produce templated content that has no depth or personalized expression (Interviewee 05).
4.1.3. Regulatory and Policy Environment for AI Journalism
In the news environment of China, the application of AI is subject to strict regulation, the government sets the direction of the development model of AI in the news industry and the direction of the content controllability and legality. This regulatory system not only affects the process of generating AI news, but also affects the way in which news organizations respond to it.
First, government regulation and censorship of news content is the most critical factor in AI news production. AI news content has to conform to national policies and news on certain subjects may be off-limits to avoid the spread of information that is against the government’s stand (Interviewees 07, 11). Furthermore, AI-powered content review systems are used for public opinion monitoring to ensure that news reports are in conformity with the socialist core values and prevent the undermining of social harmony (Interviewee 07). In regard to data security and privacy protection, the government has set higher standards for news organizations to avoid the misuse of news data by AI and to prevent the possible leakage and misuse of information (Interviewee 06).
Second, the usage of AI news tools is prohibited in China, particularly in the use of transnational AI technologies. News organizations cannot apply certain foreign AI tools (e.g. Due to policy and regulatory barriers, certain foreign AI tools such as ChatGPT are prohibited from use in the news industry and news organizations have to be cautious when selecting AI technologies that are compliant (Interviewee 10). On the contrary, Chinese news organizations tend to utilize domestic AI tools such as Deep Seek and Beanbag to guarantee that the technology is manageable and that it meets the country’s data security requirements (Interviewee 11).
Last, to respond to this policy environment, the news industry has employed several coping mechanisms. News producers of AI news ensure that the content they create is legal and safe to avoid being audited and to guarantee that news content does not violate national regulations (Interviewee 11). In high-risk areas, for instance, in the case of medical news and public policy, AI can be used in conjunction with human evaluation to minimize the chances of making a mistake and to enhance the credibility of the content (Interviewee 03).
AI news development in China requires a balance of policy regulation and technological innovation. On the one hand, the Chinese government has strictly limited the tools and content of AI news, while on the other hand, the government has provided new technological support to the news industry through investments in AI infrastructure and the promotion of local AI technologies. To achieve this, the development of AI news should occur within a compliance framework and discover how to use the innovative potential of AI within the policy boundaries to improve the efficiency and quality of news production.
4.1.4. Future Trends of AI News
In response to the future development of AI in the news industry, interviewees suggested a number of key directions for improvement to ensure that AI news can meet the requirements of truthfulness, fairness and compliance while improving efficiency.
Firstly, in terms of technological optimization, AI needs to further enhance the authenticity of content, reduce the phenomenon of hallucinations, and improve the ability to understand the context of the news to ensure the accuracy and credibility of news information (Interviewee 06). In addition, optimizing the training data of AI and reducing algorithmic bias to make news content more impartial and diverse is another major challenge for the development of the technology (Interviewee 08). At the same time, AI recommendation systems should be improved to reduce the filter bubble’s effect and ensure that users have access to a more diverse range of news content, thus increasing the openness and inclusiveness of information (Interviewee 07).
Secondly, at the level of industry development, AI’s cross-cultural communication ability will become an important driving force for news globalization, and AI’s multi-language translation technology can optimize the dissemination of international news, making it easier for audiences from different language backgrounds to access news information (Interviewee 06). In addition, AI can help optimize the news production process and reduce repetitive work, so that journalists can focus on in-depth investigations and analytical reports, enhancing the professionalism and impact of news (Interviewee 05). At the same time, AI can also be used to enhance user interaction experience, for example, through interactive news and intelligent news assistants, enabling readers to access and understand news in a more personalized way, improving news interest and user stickiness (Interviewee 07).
In the direction of policy and ethics, the development of AI news requires the establishment of clearer ethical standards to specify the roles and responsibilities of AI in the journalism industry and to ensure the credibility and social responsibility of news reporting (Interviewee 10). Data security and privacy protection are also a critical issue in the development of AI news, and the data management system needs to be further optimized to prevent AIs from misusing news data and to ensure the safety of user information (Interviewee 06).
The future development of AI news should be centered around “authenticity, fairness, and compliance.” Technological optimization should enhance the reliability and accuracy of news content; industry development should promote the responsible integration of AI into news production; and improvements to policy and ethical frameworks will provide the foundation for the sustainable growth of AI journalism. In this process, exploring effective models of human-computer collaboration will be key, allowing AI to contribute its efficiency advantages without compromising the professionalism and credibility of the journalism industry.
The future development of AI news should be centered around “authenticity, fairness and compliance”. Technological optimization should ensure the authenticity and reliability of news content, the development of the industry needs to promote the rational application of AI in news production, and the improvement of policy and ethical frameworks will provide the necessary guarantee for the sustainable development of AI news. In this process, the exploration of human-computer collaboration mode will be key, so that AI can take advantage of its efficiency without compromising the professionalism and credibility of the news industry.
After selective coding, this study draws the following conclusions:
1) AI is reshaping China’s journalism industry, improving the efficiency of news production, and driving news content diversification.
2) AI news faces authenticity, bias, and ethical issues, which have not yet been adequately addressed and still require improvements in technology and industry standards.
3) The development of AI news in China is strictly regulated, and the policy environment both restricts the free application of AI and promotes the compliant innovation of AI in the news industry.
4) Future development should focus on “truthfulness, fairness, and compliance”, and make AI play a more positive role in the news industry through technical optimization and policy adjustment.
4.2. Theoretical Coding
Building on selective coding, theoretical coding was conducted to integrate all core categories and construct a systematic theoretical model explaining the development logic and impact mechanisms of Artificial Intelligence (AI) in China’s journalism industry. This study developed the Embedded AI Journalism Model (EAJM), which reveals the core mechanisms of AI development in Chinese journalism by highlighting the interactions among three core variables: technology-driven forces, regulatory constraints, and ethical adaptation. Together, these factors shape the functional roles, scope of influence, and future trajectories of AI in news production.
Specifically, the EAJM model consists of three core components: technology-driven, regulatory constraints and ethical adaptation. Among them, “technology-driven” emphasizes AI’s advantages in innovation and efficiency. AI significantly accelerates news production, enriches content diversity, fosters industry innovation, and supports the ongoing optimization of human-computer collaboration. “Regulatory constraints” refers to the government’s strict oversight of news content to ensure that AI-generated news aligns with legal standards and ideological directives. Regulation also indirectly encourages news organizations to adopt domestic AI tools (e.g., Deep Seek, Beanbag), reducing dependence on foreign technologies “Ethical Adaptation” is about how journalists and news organizations deal with ethical issues in the use of AI. It also emphasizes the need to protect the authenticity, transparency and accountability of AI use and to establish an ethical code that is suitable for the Chinese media context.
The model further highlights the dynamic interaction between these three core variables. When innovation based on technology is paired with rules and regulations, the outcome is increased efficiency with content that is politically and legally sound. The relationship between regulatory constraints and ethical adaptation increases the legitimacy, credibility, and transparency of AI-generated content. The harmony between technology-driven development and ethical adaptation enhances the effectiveness of human-computer collaboration, enhancing the quality of journalism and ensuring ethical standards.
In conclusion, the EAJM theoretical model presents the intricate relationships between technological advancement, legal frameworks, and moral obligations. It systematically explains the embedded development of AI in China’s journalism industry and its practical implications, and provides a sound theoretical framework and analytical framework for future research.
5. Discussion and Conclusion
5.1. Interpretation of Findings
Through open coding, axial coding, and selective coding, this study summarizes the core logic and impact mechanisms of AI technology in China’s journalism industry. AI has significantly improved the efficiency and diversity of news production and promoted innovation in news dissemination. However, it has also raised new challenges related to authenticity, algorithmic bias, and professional ethics, which require both technological optimization and improvements in industry standards. In addition, the strict regulatory environment partially restricts the free application of AI technologies but also motivates news organizations to more actively adopt localized AI tools in pursuit of compliant innovation.
Further, the Embedded Development Model of AI in Chinese Journalism (EAJM), constructed through theoretical coding, reveals the interaction mechanism among three core variables: technology-driven, regulatory constraints, and ethical adaptation. These variables collectively influence the functional roles, communicative impacts, and future trajectories of AI in the news industry. The model determines the main difficulty for AI journalism to progress which involves managing technological efficiency against news reporting requirements for truthfulness fairness and social responsibility. The model establishes both theoretical and practical guidelines for the sustainable growth of AI journalism.
5.2. Comparison with Other Studies
The EAJM framework developed in this research shows both similarities and distinct features when compared to previous research findings. The research of Sun et al. (2024) confirms our study’s observation about AI benefits for Chinese journalism through increased production efficiency and content innovation and public approval of AI technology applications. The research differs from Sun et al. because it examines the complex relationship between technology and regulatory and ethical considerations.
Moreover, in line with the findings of Ji et al. (2024) and Jamil (2023), this study emphasizes the ethical and regulatory challenges of AI journalism within authoritarian or competitive authoritarian contexts. For example, similar to Ji et al., this research finds that Chinese journalists often navigate a dual role—serving as watchdogs while simultaneously supporting national interests—which presents complex professional and ethical challenges. Likewise, this study agrees with Jamil’s assertion that AI-automated journalism may pose legal and ethical threats in restrictive media environments, thereby jeopardizing media freedom.
However, this study distinguishes itself by being the first to propose and construct an embedded theoretical model (EAJM) that encompasses the interactive dynamics of the three variables: technology-driven development, regulatory constraints, and ethical adaptation. This model offers a comprehensive analytical framework for systematically understanding AI’s unique logic of development and its application in the Chinese journalism context—representing the core contribution of this paper.
5.3. Limitations
This study has some limitations in its analysis of the application and impact of AI technology in China’s journalism industry. Firstly, the research employed qualitative methodology through in-depth interviews, using primary data from a relatively small number of participants. Most of the participants were either journalism and communication professionals or students, which may reduce the external validity of the findings. However, this study is exploratory in nature, aiming to generate initial insights rather than achieve statistical generalizability. Participants were selected for their proximity to AI use in media contexts, offering relevant and experience-based perspectives.
Secondly, the results obtained are dependent on the individual experiences and cognitive processes of the respondents which may bring in bias and subjectivity to the results. Future studies could adopt quantitative or mixed-methods designs to test the broader applicability of these findings.
Furthermore, the study only concentrates on journalism in Chinese mainland and does not consider the other regions of Hong Kong, Macao, and Taiwan, or other cultural contexts. This makes it difficult to give a more global view of the implications of AI journalism. Last but not the least, as AI technologies and policies related to them are changing at a fast pace, the EAJM model proposed in this study may not fully capture the current or potential future dynamics.
5.4. Future Research Directions
Future research can be carried out in depth around the following aspects: first, the diversity and scale of the research samples can be increased, and the performance and impact of AI in different news environments around the world can be more comprehensively explored through cross-regional and cross-cultural comparative studies, especially comparing the similarities and differences between Chinese mainland and Hong Kong region, Macao region, Taiwan region, and other countries and regions; second, consideration can be given to combining qualitative and quantitative research methods to further validate this study’s proposed model by means of large-scale questionnaire surveys and experimental research to further validate the theoretical model (EAJM) proposed in this study and improve the robustness and generalizability of the findings; furthermore, future research should pay attention to the differences in the acceptance of AI news among different audience groups, especially the differences in the cognition and trust mechanisms of AI news content between younger and older groups. In addition, with the continuous development of AI technology and policy environment, long-term tracking and research on the dynamic evolution of the application of AI technology in the field of journalism and its social impacts will contribute to a more in-depth understanding of the interaction mechanism between AI and journalism. Finally, researchers can further explore how to establish ethical standards and regulatory policies for journalism adapted to the AI era, to ensure the truthfulness, fairness and compliance of AI journalism, and promote the sustainable development of AI journalism.