A study of the perception and attitude of Chinese users to media content created by generative AI

Bulgarova Bella Akhmedovna ORCID: 0000-0001-6005-2505 PhD in Philology Associate Professor; Department of Mass Communications; RUDN University Miklukho-Maklaya str., 6, Moscow, 117198, Russia bulgarova-ba@rudn.ru Other publications by this author     Ma Fei ORCID: 0009-0002-5688-1348 Postgraduate student; Faculty of Philology; P. Lumumba Peoples' Friendship University of Russia Miklukho-Maklaya str., 6, Moscow, 117198, Russia ma_fei@mail.ru Other publications by this author     Chen' Fenglan ORCID: 0009-0007-2297-6126 Postgraduate student; Faculty of Philology; P. Lumumba Peoples' Friendship University of Russia Miklukho-Maklaya str., 6, Moscow, 117198, Russia fenglan.chen@mail.ru Other publications by this author

Fan' Zekun ORCID: 0009-0001-3104-1580 Graduate student; Department of Mass Communications; P. Lumumba Peoples' Friendship University of Russia Miklukho-Maklaya str., 6, Moscow, 117198, Russia fan_ze@mail.ru

Introduction

Rapid technological iteration has always been a major factor in media development. New technologies are key elements and supporting forces of profound changes in the media industry, and artificial intelligence has become a new strategic window of opportunity for deep integration and development of media ^{[1, pp. 9-11]}, contributing to a paradigm shift in the production of media content and communication ^{[2, pp. 94-104]}. Thanks to the rapid development of technology, the transformative power of generative AI has attracted widespread attention from academia. In current research, some scientists argue that the rapid development of generative AI promotes innovation in media application ecosystems ^{[3, pp. 13-20]}, providing unlimited opportunities for the media industry through human-computer interaction models ^{[4, pp. 105-113]}. The rapid development of generative artificial intelligence technologies has contributed to the deep integration of the media business and artificial intelligence, stimulated the digital transformation of media driven by generative artificial intelligence, and increased the efficiency of information processing and communication ^{[5, pp. 50-55]}. Generative artificial intelligence acts as a catalyst for the intellectual development of the media industry, rapidly bringing technologies and functions that have been studied in the media field for a long time, but have not yet achieved ideal results, to the level of industrial application ^{[6, pp. 9-13]}. At the same time, some scientists believe that, from the point of view of user cognition, generative AI content changes the structural mechanism of user experience and carries numerous cognitive risks ^{[7, pp. 56-63]}. Some scientists have also drawn attention to gaps in the confidentiality of the information generated, arguing that there are ethical risks in applications with generating artificial intelligence, such as: violation of personal integrity, inaccurate content and transmission of false information ^{[8, pp. 39-42]}, algorithmic bias ^{[9, pp. 60-68]}, the illusion of AI ^{[10, p. 53-63]}, the crisis of social trust ^{[11, pp. 130-139]}, thereby affecting the foundations of trust in mass communication ^[12]. Other researchers consider the technology of generative AI from the point of view of management and analyze data, algorithms, platforms and users, offering reliable principles of ethical management ^{[13, pp. 35-42]} and creating a system of AI ethics tools ^{[14, pp. 79-90]} and a regulatory system ^{[15, pp. 108-115]}.

To date, the main pool of relevant research on the issue under consideration has focused on the performance spikes and increased creativity caused by generative AI in the production of media content and communication, the ethical risks and challenges associated with generative AI technology, as well as various management measures proposed to manage AI in the age of intelligent communication. However, there is relatively little scientific literature devoted to the study of cognitive abilities and attitudes of users, which are key in the process of transmitting media content. To fill this gap in research, this article focuses on the study of subjective perception and user attitudes towards media content generated by generative AI.

Research methodology

To study the perception and attitude of users to the generated content, a survey method was used among Chinese users (522 questionnaires). The collected data was then analyzed to process and summarize the results of the study.

The study includes the following stages:

1. Review of theoretical literature and data collection: first, the literature on the research topic is collected and systematized, the development of media content created by generative AI is analyzed;

2. Research planning and data analysis: based on the research topic, a questionnaire was developed and data was collected;

3. Discussion of the survey results and conclusions;

Research design and data analysis

Questionnaire design

The authors of the article have developed a questionnaire to better understand the perception and attitude of users to media content generated by artificial intelligence. The questionnaire mainly included modules on basic information, habits and experiences of using generative AI, influence and ethics.

The following questions were asked to the Chinese user audience:

1. Have you ever encountered intelligent content products or services from media organizations?

2. What are the main scenarios in which you encounter media content products or services created using artificial intelligence?

3. What do you think is the main driving force behind generative content created using artificial intelligence?

4. In what aspects do you think products or services with content generated by generative artificial intelligence work best?

5. What is your general attitude towards generative content created using artificial intelligence?

6. What problems do you think there are with generative content created using artificial intelligence?

The questionnaire was distributed online using the Wenjuanxing system, and 522 valid questionnaires were collected.

Based on the basic information contained in the samples obtained, the respondents were mainly between the ages of 18 and 40: 35.63% were aged 18-30 and 45.02% were aged 31-40. There are slightly more women (53.26%) than men (46.74%). Regarding education, 59.96% had a bachelor's degree, indicating a wider acceptance of AI-generated content among more educated people. Regarding the location of users, 33.52% chose "first-level cities (such as Beijing and Shanghai)", while 40.42% chose "second-level cities". Data analysis shows that people living in big cities are more likely to become familiar with AI-generated media, and highly educated and younger users are more supportive of AI-generated media.

Data analysis and discussion

An analysis of the collected sample data showed that in response to the question "Have you ever encountered intelligent content products or services of media organizations?" (see Figure 1), 34.67% of respondents chose the option "I often use them and find them very convenient," while 43.87% chose the option "I I use them from time to time, and their impressions are so-so." These data indicate that almost 80% of respondents actively or passively used media content products or services created using generative artificial intelligence, and rated them from medium to high. It is assumed that users generally have a positive attitude towards this kind of media content.

Figure 1. Have you ever encountered intelligent content products or services from media organizations?

Regarding the question: "What are the main scenarios in which you encounter media content products or services created using artificial intelligence?" (see Figure 2), 61.11% of respondents chose "receiving daily news and information"; 66.28% chose "observing specific events (such as sporting events, political events etc.)"; and 51.34% chose "social media communication (exchange of interesting news or opinions)". Based on the data obtained, it can be argued that generative media content created with the help of artificial intelligence has already invisibly and comprehensively penetrated into people's daily lives. Users use media content created with the help of artificial intelligence in various scenarios, which makes it an important part of their lives. Users receive news and information from media content created using artificial intelligence and share interesting stories.

Drawing 2. What are the main scenarios in which you encounter media content products or services created using artificial intelligence?

Regarding the question: "What do you think is the main driving force behind generative content created using artificial intelligence?" (see Figure 3), 45.59% of respondents chose "technological progress (artificial intelligence, big data, etc.)", and 46.74% chose "changes in audience needs (for example, changes in demand for personalized and diverse content)." Since the beginning of human civilization, technological development has been the most important driving force of human progress, as well as the main logic and driving force behind the development of mass media, leading to changes in content production and disrupting communication methods. The constant updating of technologies leads to the fact that the evolution of media is more and more consistent with the physiological characteristics of human communication and information transmission, corresponds to the methods of human communication and constantly meets the needs of people in a variety of content. The history of the development of mass media also shows that the effectiveness of all media is becoming more and more human-oriented, meeting people's real needs for communication methods and sensory experiences. All mass media are trying to find their own niche ^{[16, p.118]}. Generative artificial intelligence constantly meets the needs of users and finds its own "human niche" through continuous technological integrations.

Figure 3. What do you think is the main driving force behind generative content created using artificial intelligence?

Regarding the question "In what aspects do you think products or services with content generated by generative artificial intelligence work best?" (see Figure 4), 74.14% of respondents chose "Multimedia content presentation formats (such as video, audio, animation, etc.)"; 66.67% chose "We are very satisfied with the accuracy of personalized content recommendations"; and 45.98% chose "Interactivity (for example, commenting, liking and information sharing functions)". The data shows that people are most satisfied with multimedia content presentation formats (such as video, audio, or animation). Thanks to the widespread use of generative artificial intelligence models such as ChatGPT, Sora, Deepseek and Seedance, the media not only have the opportunity to overcome labor and time constraints, but also to be deeply embedded in the production process of media content and communication, which significantly increases the efficiency and creativity of media content production and provides users with richer presentation formats and a more intelligent sensory experience.

Figure 4. In what aspects do you think products or services with content generated by generative artificial intelligence work best?

As for the question: "What is your general attitude towards generative content created using artificial intelligence?" (see Figure 5), the most popular option was "I support it to some extent, but I have some concerns," which amounted to 47.51%; the most popular option was "I strongly support it, this is an inevitable trend in development Mass media", which amounted to 38.31%. In addition, 11.88% chose a "Neutral, wait-and-see attitude," while 2.3% chose a "Not very supportive one, concerned about potential problems." These data show that almost 40% of users strongly support generative media content created using artificial intelligence, believing that this is an inevitable trend in intelligent media transformation. Almost half (47.51%) of users also have a more objective and rational attitude towards media content created using artificial intelligence. This indicates that the majority of users approve of content created using artificial intelligence. However, they also recognize that while generative artificial intelligence improves the efficiency of media content production and communication by providing users with a more personalized and intelligent sensory experience, they are also aware of the potential risks involved. Therefore, while they support the development and application of generative AI in the media field, they also have some concerns about the potential risks this could entail.

Figure 5. What is your general attitude towards generative content created using artificial intelligence?

Regarding the question: "What do you think are the problems with generative content created using artificial intelligence?" (see Figure 6), 30.65% of respondents chose "Excessive technical costs"; 65.33% chose "Data privacy and security issues"; 49.81% chose "Algorithmic bias and ethical risks"; 52.68% chose "Intelligent recommendation algorithms can lead to an information cocoon effect"; 43.3% chose "Information Security". overload, which makes it difficult to filter out really valuable content"; and 28.93% chose "Spreading false information". It can be seen that people in general are concerned about the potential problems associated with the media content generated by artificial intelligence. Among these problems, the most significant are the problems of confidentiality and data security ^{[17, pp. 93-99]}, the potential effect of the information cocoon caused by intelligent recommendation algorithms ^{[18, pp.11-16]}, as well as algorithmic bias and ethical risks. In the context of the rapid development of intelligent technologies, data, as the basis of technological functioning, is increasingly demonstrating its value and importance in intellectual development, which makes data privacy and security the main issue and subject of discussion. According to the "Data Leak Trends Analysis Report for 2024", the share of incidents related to data leaks showed an overall upward trend in each month of 2024, with simple data leaks accounting for up to 66% ^[19]. These data also show the growing importance of data privacy and security risks, which have become a significant factor influencing people's perceptions. Along with the continuous updating and integration of various digital technologies, algorithmic recommendations based on personal preferences accelerate the formation of information cocoons. Although algorithmic bias at first glance allows users to choose content based on their personal interests, it also inadvertently blocks valuable information that may be of interest to users or goes beyond their interests ^{[20, pp. 42-50]}. These trends will inevitably lead to a uniformity of information sources, and as a result, a uniformity of user opinions.

Figure 6. What problems do you think exist with generative content created using artificial intelligence?

Conclusions

Generative content created with the help of artificial intelligence is an inevitable trend in the development of media. From the era of oral communication to the era of online communication, technological innovations have always been the most important driving force behind the development of media. The rapid development of generative artificial intelligence means that AI technology is gradually moving beyond its function as a simple tool and is beginning to move from quantitative to qualitative changes ^{[21, pp.25-35]}. In particular, the widespread adoption and application of generative artificial intelligence models such as ChatGPT, Sora, Deepseek, and Seedance are changing the development ecosystem of the entire media industry from different perspectives, including production and communication efficiency, models, and user experience. With thematic planning, generative AI can automatically create thematic libraries that can be referenced by media professionals. When creating content, it can help with creating initial drafts of text and images, editing videos, and multilingual translation. For example, the short drama "Chinese Mythology", launched by CCTV based on artificial intelligence, was created entirely using artificial intelligence, including video and voice acting, which significantly reduced production costs. Thanks to the rapid implementation of generative artificial intelligence, he will act as a partner in collaboration with people deeply involved in the production of media content and communication. This not only significantly increases the efficiency of production and communication, but also enriches the forms of media content transmission, significantly increasing opportunities and creativity in the production of media content. Generative content created with the help of AI is an inevitable trend in the transformation and intellectual development of the ecosystem of media communications.

It is also necessary to rationally assess the risks associated with generative artificial intelligence, since it has become a powerful force that creates huge opportunities in the field of mass media and provides a leap in productivity. In the process of information communication, deep fakes, a product of the development of artificial intelligence technologies, blur the boundaries between reality and fiction, providing technical support for the spread of misinformation and exacerbating the prevalence of misleading the public. Intelligent algorithms can imperceptibly increase bias and discrimination; users may inadvertently enter confidential information when interacting with AI, which creates risks of data privacy violations. These risks challenge not only the core values of the media industry, but also its healthy development. Faced with the surge in productivity and ethical risks associated with generative artificial intelligence, we should neither accept it unconditionally nor reject it completely. Instead, we must take a rational and prudent position, on the one hand, carefully monitoring the risks associated with generative artificial intelligence, and on the other, looking for solutions in the field of innovative development and finding a dynamic balance between innovative development and security control. Only by maintaining rational and vigilant awareness of these risks is it possible to increase the benefits and minimize the harm caused by generative AI, ensuring that technology truly serves human development.

Generative artificial intelligence is transforming from a single auxiliary tool into a media infrastructure, changing media content production and communication, while creating risks such as data privacy and misinformation. To eliminate these risks, it is necessary to create a systematic and collaborative comprehensive management structure for generating AI, both technologically and fundamentally, in order to better promote its healthy development. From the point of view of technological management, on the one hand, a data quality assessment process should be created to monitor the accuracy and integrity of data; on the other hand, encryption and anonymization technologies should be introduced to ensure the security of personal data ^{[22, pp. 73-85]}. Meanwhile, in order to eliminate the risks associated with algorithmic black boxes and the illusions of AI inherent in generative AI, it is necessary to constantly increase the transparency and explainability of algorithms. From a management perspective, first, the government should formulate rules for the application and management of AI, strengthen oversight, and ensure effective enforcement of laws and regulations through interagency cooperation. Secondly, platforms must integrate management requirements into their business processes and fulfill their obligations by strengthening technological capabilities. Third, users should actively improve their digital literacy by increasing their ability to recognize the authenticity of content and identify risks. The creation of an effective AI management system is aimed at ensuring a balance between innovation and safe development. This not only fully exploits the capabilities of technology, but also effectively manages potential risks and challenges in AI development, jointly creating a secure and reliable information environment.

The article is published in the version approved by the reviewers (after receiving a positive review recommending the manuscript for publication) with corrections made by the author (after receiving the editor’s comments, if any).
Read all reviews on this article