What is AI and what is AIGC?
I believe that artificial intelligence is not a new word, if we check the Wechsler dictionary, we will find that it defines AI as “a branch of computer science dealing with the simulation of intelligent behavior in computers” or “the capability of a machine to imitate intelligent human behavior.” But some scholars argue that AI is not just a collection of technologies and algorithms, but a complex phenomenon involving a wide range of social, economic, and political forces (Crawford, 2021). This definition challenges the perception of AI as an independent, autonomous intelligence and emphasizes how the development and deployment of AI reflect and reinforce existing power dynamics, biases, and inequalities.
So, what is AIGC? When it comes to AIGC, UGC and PGC have to be mentioned. Simply put, we have gone through three eras Web 1.0, Web 2.0, and Web 3.0 era. In the Web1.0 era, the content production method is Professional Generated Content (PGC), PGC refers to the content created, edited, and published by professional content creators or teams, and this creation method originated from the traditional media era, such as newspapers, magazines, TV and movies. In the Web 2.0 era, the content production method is User Generated Content (UGC), UGC refers to the content created, edited, and published by ordinary users or audiences, and this kind of production method is popular with the emergence of social networks and blogs, etc. The application scenarios of UGC include social networks, online forums, blogs, knowledge-sharing platforms, etc. UGC can be images, videos, and other content. UGC can be in the form of pictures, videos, music, blogs, comments, etc. It can be individual, group, or even global. Representative communities or applications are Facebook, Twitter, Instagram, etc. In the era of Web 3.0, the content production method is Artificial Intelligence Generated Content (AIGC), AIGC is an emerging creation that uses artificial intelligence technology and natural language processing technology to generate content. Currently, AIGC is mainly used in four categories: writing, audio, image, and video. Among them, the writing tools are mainly OpenAI’s GPT series, which can generate relevant articles, news reports, stories, dialogues, and even codes according to the input instructions (Translated by Content Engine, L. L. C., 2023). Other tools in the writing category are NewBing, Elephas, WordAI, etc. In the field of music creation, AIVA (Artificial Intelligence Virtual Artist) can generate original musical compositions based on user input, and also supports collaborative creation with human musicians. Other audio tools include Fliki, an AI dubbing tool. In the image creation field, Midjourney and Stable Diffusion can generate image works based on user input. In addition, the image field has catalyzed many subdivision tools, such as ProfilePicture.AI and Astria for generating avatars; interior AI, and AI Room Planner for room design, and so on. In the field of video creation, AIGC is mainly used in video editing and special effects, such as Google’s DeepMind’s AI-assisted video editing tool Kinetics, which will be launched in 2019, and Adobe’s Sensis.Other video tools include Vidyo, Tavus, etc.
| Internet Form | Web1.0 | Web2.0 | Web3.0 |
| Content production method | PGC | UGC | AIGC |
| Subject of production | Professional media | Non-professional | AI |
| Characteristics | High quality | Rich in content | High production efficiency |
As mentioned earlier, in the digital age, PGC will continue to leverage its expertise and maintain high quality and depth of content but will struggle to scale and reach people quickly. UGC will continue to exist as a means of personal expression and community sharing but will need to focus on the quality of content and copyright protection. AIGC, on the other hand, will increasingly be used for rapid content generation and automated production but will need to focus on algorithmic accuracy and model training. AIGC will be increasingly used for rapid content generation and automated production but needs to focus on algorithmic accuracy and model training.
How to see AI as a double-edged sword
AI has been controversial since its birth. As Liu et al. (2023)state, the development of AI has brought many opportunities for society that have never been seen before, but it is also accompanied by a series of risks and challenges.
Positive impacts
The development of AI technology promotes efficiency and productivity. This is because AI is able to automate complex tasks and can effectively improve work efficiency, especially in the areas of data processing, manufacturing, and services. Second is the role of AI in facilitating innovation and technological development. Artificial Intelligence plays an important role in many industries such as healthcare, finance, and transport, and has already solved many traditional technical difficulties well by using methods such as deep learning and big data analysis. At the same time, AI can also improve the quality of life. For example, the application of AI technology in home management and automation makes daily life easier. Smart home systems can automatically adjust indoor temperature, lighting and security systems according to the habits of the occupants. This not only improves living comfort, but also helps to save energy and reduce emissions.
Negative Impacts
Firstly AI affects unemployment and job market changes. As a result of AI development, automation can cause a drop in demand for specific industries, affecting workers with lower skill levels, and thus further aggravating social inequality. The second point is the issue of privacy and surveillance (PASQUALE, 2015). Advances in AI enhance data collection and processing, and while this data can be used to improve service quality and security, it is also highly susceptible to misuse. It can also violate individuals’ right to privacy, especially in terms of surveillance and data analysis. The third point ethical and moral issues are difficult to deal with. AI often shows some limitations in dealing with scenarios involving moral and ethical issues. This is mainly because AI systems usually rely on data-driven algorithms that lack the emotional complexity and social awareness that humans rely on for moral judgement and ethical decision-making. So it is difficult for AI to have integrated emotional and social awareness of judgement like humans.
How to balance the relationship between technological progress and social ethics?
Balancing the relationship between technological advancement and social ethics appears to be crucial in the field of AI. Firstly, it is fundamental to establish and strictly follow AI ethical guidelines, which should include key aspects such as data privacy, transparency, fairness, and responsibility. For example, PASQUALE (2015) shows that AI systems must respect the privacy of users when handling data and ensure transparency and understanding of their decision-making processes to increase public trust. Second, increasing diversity and inclusion in the AI development process is critical to reducing bias. This involves incorporating members of different genders, races, and cultural backgrounds into the team composition and ensuring that the data used for algorithm training covers a wide range of topics to prevent injustices due to data bias. Further, with the rapid development of AI technology, ongoing regulation, and evaluation are particularly important, and governments and relevant regulators need to develop standards and frameworks to oversee AI applications to ensure that they do not infringe on individual rights or cause social instability. It is also essential to provide ethics education and training for AI developers and users, enhancing their awareness of potential ethical issues through education and developing the habit of considering ethics when designing and implementing AI solutions. Finally, given the global impact of AI technologies, international cooperation coordination and standardization are also vitally important, and global ethical challenges can be addressed more effectively by coordinating national policies and standards through international organizations and forums.
What signals have been released by China’s largest official media CMG’s introduction of norms for the use of AI in media?
With the rapid development of AI big models and all kinds of text generation, text-to-graphics, and text-to-video tools this year, the global media industry is also ushering in an all-around change in the way content is produced, disseminated, and consumed.
To apply AI in a more standardized, reasonable, safe, and efficient way, on 21 March, CCTV formally formulated and introduced the “CCTV Artificial Intelligence Usage Specification (for Trial Implementation)”, which is China’s first standardized standard for the use of AI in the media. The introduction of this specification encourages the innovative application of AI, and the use of new quality productivity to promote high-quality development and lead the development of the industry. On 27 March, CCTV held a new AI product conference at the Boao Forum for Asia, where several new AI products produced by the CMG to interpret the Chinese antiquities collection and Chinese classical myths met with the audience.
A day after the launch of the AI channel on the main station, OpenAI broke the news that its innovative text-to-video generator, Sora, would soon enter Hollywood, a development that has attracted a great deal of attention from the global film industry. OpenAI is expected to meet with many top Hollywood studios, talent agencies, and media executives in Los Angeles to discuss potential opportunities for cooperation.
Along with Sora’s explosion out of the circle, combined with the demonstration and driving effect of the main station, China’s local media, film, and television practitioners around the layout of AI continue to speed up. A variety of AIGC tools, products, and application scenarios are coming out. From the platform, in February this year, the head office CCTV took the lead in setting up an AI studio, driving the layout of broadcasting at all levels across the country on AI. Shanghai, Beijing, Chengdu, and Henan TV stations followed, relying on their advantages, accelerating their development, occupying the AIGC market, and promoting the integrated development of the production, study, research, and use of AI technology in the media field through a diversified mode of cooperation.
In terms of works, AIGC technology, with its powerful content generation capability, opens up diversification of content innovation and provides broadcast media with a new way of content creation and presentation. Broadcasters at all levels have accelerated AIGC content creation, transforming the media’s new quality productivity into AIGC works such as “Ode to a Thousand Autumn Poems” and “Chinese Myths”, which promote traditional culture, and “On the Trail of the Chinese Dragon”, which tell a good Chinese story. The innovative exploration of the application of AIGC technology in the media field has been transformed into the concrete practice of seizing the high ground of future industrial competition to empower the cultural industry and international communication.
CMG’s introduction of specifications for the use of AI in the media and the active participation of local media, film, and television practitioners in the AIGC industry is the media sector’s positive response to the technological revolution and changes in market demand. This is not only a technological innovation, but also a comprehensive change in the way of content production and consumption, heralding the reshaping and leap of the media industry in the wave of intelligence.
How to Governance
The key to dealing with the double-edged sword attributes of AI lies in how to balance its pros and cons and formulate appropriate policies and regulatory measures. Governance of AI requires a comprehensive strategy that includes legislation and policy development, establishment of ethical standards, technology vetting, education and training, international cooperation, and public participation. First, national and international institutions need to formulate strict laws and policies to regulate the development and application of AI, ensure data protection, privacy, and security, and strengthen the transparency and fairness of AI systems (Huang et al., 2023). Second, AI development must comply with ethical principles, such as fairness, transparency, and accountability as emphasized in the EU’s Ethical Guidelines for AI (EU unveils ethics guidelines for Artificial Intelligence, 2019), in order to ensure the legitimacy of the technology as well as its morality. It is also crucial to conduct regular technology audits and assessments to detect the safety, efficacy, and potential bias of AI systems through algorithmic reviews and fairness assessments of datasets.
Education and training are also integral to AI governance, as the public and policymakers’ understanding of AI technologies should be enhanced, and AI developers should be provided with the necessary ethical education to increase awareness of AI’s potential and risks. International cooperation is also extremely important. Given the global impact of AI technologies, the joint development of standards and sharing of best practices through international organizations and multilateral forums can effectively coordinate AI policies and applications on a global scale. Finally, it is crucial to ensure transparency and public participation in AI projects. Enabling the public to understand AI applications and participate in regulatory discussions can help increase public trust and acceptance of AI technologies.
In sum, the challenge of governance of AI lies in the fact that rapid technological development usually makes legal and policy development lag behind, and thus continuous research, interdisciplinary cooperation, and flexible policy adjustments are needed to address this challenge.
References
Crawford, K. (2021). Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. New Haven: Yale University Press. https://doi.org/10.12987/9780300252392
EU unveils ethics guidelines for Artificial Intelligence. (2019, Apr 08). AFP International Text Wire in English https://www.proquest.com/wire-feeds/eu-unveils-ethics-guidelines-artificial/docview/2204524332/se-2
Huang, K., Wang, Y., Zhu, F., Chen, X., & Xing, C. (Eds.). (2023). Beyond AI : ChatGPT, Web3, and the business landscape of tomorrow (1st ed.). Springer.
Liu, Y., Fu, Z., Li, T. (2023). How Can Artificial Intelligence Transform the Future Design Paradigm and Its Innovative Competency Requisition: Opportunities and Challenges. In: Degen, H., Ntoa, S., Moallem, A. (eds) HCI International 2023 – Late Breaking Papers. HCII 2023. Lecture Notes in Computer Science, vol 14059. Springer, Cham. https://doi.org/10.1007/978-3-031-48057-7_9
PASQUALE, F. (2015). The Black Box Society: The Secret Algorithms That Control Money and Information. Harvard University Press. http://www.jstor.org/stable/j.ctt13x0hch
由 Content Engine, LLC 翻译（2023 年 4 月 10 日）。打开AI GPT Chat，问题很多，答案很少。 CE Noticias Financieras https://www.proquest.com/wire-feeds/open-ai-gpt-chat-many-questions-few-answers/docview/2799445108/se-2
Be the first to comment