Abstract The breakthrough applications of artificial intelligence have brought about significant historical social changes and have presented a new context for discussing the future of the humanities in China. The innovative research methods in the humanities, triggered by these changes, form an important starting point for current discussions. Based on a question-and-answer exploration of multiple logically related issues regarding “the explosion of artificial intelligence and the future of the humanities,” we can predict the groundbreaking developments of generative artificial intelligence and its impact on the humanities both now and in the future. Artificial intelligence is likely to bring significant changes to the concepts related to humanities disciplines, and the research paradigms of the humanities will also undergo important changes. For us, who are facing the future intelligent humanities world, this presents both opportunities and challenges.
Keywords Artificial Intelligence, Humanities, Social Sciences, ChatGPT
Author | Chen Yuehong Source | Journal of People’s Forum: Academic Frontier, March 2024
1. Introduction
Currently, China has entered a digital intelligent society constructed by the breakthrough applications of artificial intelligence, which is likely a social consensus. Regardless of the world, focusing solely on the vast late-developing regions of Eastern Asia where we are located, whether in cities or rural areas, the digitalization of production and life has already become a common operating method in society. Our social management, economic operation, technological and cultural education, and even daily necessities are all filled with the presence of digital intelligent applications. For most people, a life without a smartphone is almost unimaginable, even to the extent that it is difficult to move without it. Relevant statistics show that nearly half of China’s GDP is related to digital intelligence. This historic social change and the social landscape it presents form a new context and starting point for our current discussions about the future prospects of the humanities.
Since the summer seminar on artificial intelligence held at Dartmouth College in the United States in 1956, the discipline of artificial intelligence (AI) has been named and has undergone constructive development for 68 years. However, it is only in recent years that human society and governments have begun to truly recognize its disruptive impact. In 2016, AlphaGo defeated the top Go players from China and South Korea, and although this single industry application of AI was sensational, it did not create anxiety or panic. However, when OpenAI’s ChatGPT quietly went online on November 30, 2022, it immediately caused a global sensation. In 2023, the wave of generative artificial intelligence models swept through Eastern and Western societies, and the so-called “hundred-model war” to build various large models intensified, with funds flooding into accounts of companies building foundational large models, vertical application small models, and supercomputing intelligent chip production companies like NVIDIA. From the construction of foundational large models to the continuous innovation of machine learning operations, from the construction of various multimodal technology models for consumer-oriented innovative applications to the construction of data models for specialized applications across various industries, within just about a year, the application ecosystem surrounding various foundational large models and generative artificial intelligence has begun to take shape in the United States, China, and some major developed economies. The functionalities of large models, which can see, hear, speak, read, write, calculate, program, and possess powerful natural language processing and graphics and image language video interaction generation capabilities, are overwhelming.
By the end of 2023, the resignation and reinstatement drama of OpenAI’s CEO, the expectations surrounding G* (presumably Q-learning, self-learning), and GPT-5, as well as the rumors that humanity seems to have touched the threshold of AGI (Artificial General Intelligence), all indicate that artificial intelligence will accelerate changes in human life, alter the majority of industries, and even change the world. This change is occurring at a speed that even science fiction cannot keep up with. Given the impact of generative artificial intelligence on society as a whole, if one attempts to discuss the present and future of artificial intelligence and the humanities, then the significant development of generative artificial intelligence represented by OpenAI in 2023 undoubtedly constitutes the main background for discussion and response.
As for the humanities, the term itself is both familiar and vague to most people. Even scholars in the humanities may find it difficult to accurately articulate the concepts, branches, and boundaries of their discipline. This article attempts to utilize GPT-4 to conduct a question-and-answer exploration of multiple logically related issues regarding “the explosion of artificial intelligence and the future of the humanities,” while preserving GPT-4’s responses verbatim and adding the author’s supplements, corrections, comments, and extensions, aiming to create a triadic interaction and collision of inspiration among the responses from artificial intelligence, the author’s comments, and readers’ readings.
2. The Concept and Scope of the Humanities
Question: Please briefly summarize the concept and scope of the humanities?
GPT-4: The humanities (Humanities) are disciplines that study aspects of human culture, history, society, and art. This field mainly focuses on human creative expression and thought, as well as how these thoughts shape our worldview and society. The main characteristics of the humanities are an emphasis on critical thinking, analysis, and reflection. The scope of the humanities is quite broad, primarily including: 1. Literature: the study of literary works from different cultures and eras, including poetry, novels, drama, etc. 2. History: the exploration of past events, figures, societies, and cultures, and their impact on the present. 3. Philosophy: the study of fundamental questions regarding existence, knowledge, values, reason, mind, and language. 4. Art History: the study of the historical development, styles, and expressive techniques of artworks. 5. Musicology: the study of the history, theory, composition, and performance of music. 6. Linguistics: the study of the structure, development, and usage of language. 7. Religious Studies: the study of religious beliefs, rituals, myths, and their impact on society and culture. 8. Cultural Studies: the exploration of the relationships between cultural phenomena, social structures, and political forces. The humanities typically emphasize interdisciplinary learning and research, encouraging scholars to explore human experiences and creations from different perspectives and methodologies. Through the study of the humanities, we can gain a deeper understanding of the diversity, complexity, and variability of human society.
Author’s Comment: We all know that GPT stands for Generative Pre-trained Transformer, a natural language processing model based on the Transformer architecture, widely used in the field of natural language processing, including natural language generation, understanding, machine translation, and question-answering systems. ChatGPT, developed by OpenAI, is a large pre-trained language model, commonly understood as a powerful chatbot based on the GPT-3 model variant. GPT-4 summarizes the humanities as “the study of aspects of human culture, history, society, and art,” which is largely accurate, indicating that its parameter training and data scale are roughly feasible and effective, but this summary is not accurate and complete enough. If it were expressed as “the humanities are disciplines that study human language, literature, history, philosophy, art, culture, etc.,” it would be more universal. GPT-4 has basically outlined the main research objects and the basic disciplinary system and structure of the humanities in its eight classifications. Furthermore, I believe that literary studies should include “literary history, literary theory, bibliography, comparative literature, and contemporary online literature”; history, philosophy, and language studies should include their “temporal evolution, chronological progression, specialized history, and spatial narrative”; additionally, it should include world history, regional history, national history, and chronological history, as well as the history of philosophy and linguistics. Of course, we cannot blame ChatGPT for being overly demanding, as it is still in its infancy, only at version 4.0, and when it iterates to a higher version, it will undoubtedly become more mature and wise. GPT-4’s assertion that “the humanities typically emphasize interdisciplinary learning and research” is relatively advanced but also subject to discussion. Nevertheless, GPT-4’s introduction to the concept and scope of the humanities in Chinese has certainly exceeded the understanding of most people in university campuses. Therefore, when we need to learn about various disciplines and professional knowledge in the future, we can consult ChatGPT first to quickly expand our knowledge and depth, especially for those mature fields where knowledge can be organized and summarized through memory storage, ChatGPT can retrieve and present it as needed, saving us a lot of effort and helping us focus more on innovative learning and work.
3. The Development and Current Situation of the Humanities: A Global Perspective
Question: Please briefly describe the development and current situation of the humanities in human history.
GPT-4: The development of the humanities has undergone a long and complex historical process, and its current situation reflects a trend of diversification and interdisciplinarity. Here is a brief overview of the history and current situation of the humanities:
Development History. Ancient Greece and Rome: The origins of the humanities can be traced back to ancient Greece and Rome, where disciplines such as philosophy, rhetoric, history, and poetry began to take shape. Thinkers like Plato and Aristotle had a profound influence on later humanities. Middle Ages: In the Middle Ages, the humanities in Europe mainly focused on religious and theological studies. At the same time, rich traditions of humanities scholarship developed in the Islamic world and in regions like India and China. Renaissance: The European Renaissance marked an important turning point for the humanities. Scholars began to emphasize the study of classical culture, highlighting human values and individuality, which promoted the development of literature, art, philosophy, and history. Modern Era, Enlightenment: The Enlightenment of the 18th century further propelled the development of the humanities, particularly philosophy and history. 19th to 20th Century: This period witnessed the formation of modern disciplinary systems, including modern linguistics, literary theory, and art history. Post World War II: After the war, the humanities began to pay more attention to social issues such as race, gender, and class. Contemporary, Diversification and Interdisciplinarity: Contemporary humanities tend towards diversification and interdisciplinarity, including emerging fields like cultural studies, media studies, and women’s studies. Digitalization and Globalization: Digitalization and globalization have had a significant impact on the humanities, promoting research from a global perspective and the development of digital humanities.
Current Challenges. Funding and Employment Issues: The humanities face challenges of reduced funding and a tightening job market. Technological and Scientific Bias: The social and educational systems are increasingly inclined towards the fields of science and technology, leading to a decline in the emphasis on the humanities. Despite these challenges, the humanities continue to play an important role in cultivating critical thinking, cultural understanding, and moral consciousness.
Author’s Comment: GPT-4 has outlined the broad contours of the history and current situation of the humanities in less than 800 words, which is not an easy task for an average researcher relying solely on mental effort. The philosophy, rhetoric, history, and poetry of Greece and Rome, religious theology of the Middle Ages, the revival of literature and art during the Renaissance, the linguistics, literary theory, and art studies of the last 200 years, cultural studies and women’s studies post World War II, and digital humanities in the era of globalization—all these have been roughly organized by GPT-4. Moreover, GPT-4 has noted how the humanities strive to address real social issues, the plight of the humanities in the tide of economic globalization, and the essential role it plays in cultivating critical thinking and moral development. However, GPT-4’s limitations are also evident. Firstly, in fields with insufficient corpus data, GPT-4 cannot provide accurate and comprehensive answers. For instance, when discussing the history and current situation of the humanities in China, India, and the Arab world, GPT-4 is vague, merely stating that “rich traditions of humanities scholarship also developed” without providing specific content, indicating a lack of parameters related to this historical and cultural region during the large model training process.
Secondly, GPT-4’s response is somewhat formulaic. For example, regardless of relevance, GPT-4 emphasizes interdisciplinary research in its summary of the humanities, which is clearly not a necessity based on facts or the inevitable development of disciplines, but rather based on some response template. This also suggests that GPT-4 is essentially an artificial technological product that has not yet evolved to a state of independent thought. Below, I will pose deeper questions about cultural differences to GPT-4 to test its information and logical storage in this regard by increasing the difficulty.
4. The Development and Conceptual Expression of the Humanities: A Chinese Perspective
Question: Could you discuss the development and conceptual expression of the humanities in Chinese history and how it differs from and has its unique characteristics compared to the West?
GPT-4: The development of the humanities in Chinese history has its uniqueness, and its conceptual expression differs from that of the West. Here is a brief description of the development of Chinese humanities: Ancient, Confucianism, Daoism, and Legalism: The development of ancient Chinese humanities is closely related to philosophical thoughts such as Confucianism, Daoism, and Legalism. Confucian teachings particularly emphasize etiquette, morality, and political philosophy, profoundly influencing China’s education and social systems. “Six Arts” Education: Ancient Chinese education included the “Six Arts,” namely rites, music, archery, charioteering, writing, and mathematics. Early forms of humanities education emphasized practicality and moral cultivation. Medieval and Ming-Qing Dynasties: During the Song, Yuan, Ming, and Qing dynasties, literati culture flourished, with poetry, calligraphy, and painting becoming important means for literati to express personal emotions and virtues. These art forms hold an extremely important position in Chinese culture. The Imperial Examination System: The implementation of the imperial examination system further promoted the development of Confucianism and literature, with examination content mainly focusing on Confucian classics and poetry composition. Modern Era, May Fourth Movement: The May Fourth Movement in the early 20th century marked the modern transformation of Chinese humanities. Western scientific, democratic, and humanistic ideas began to influence China. Modernization of Literature and Art: Modern literature, drama, and painting began to absorb Western elements, forming unique modern expressions. Contemporary, Interdisciplinary Development: Contemporary Chinese humanities have begun to incorporate more interdisciplinary elements, such as sociology and psychology. Global Perspective: With China’s rising status in globalization, Chinese humanities have also begun to engage more in international academic dialogue and exchange.
Characteristics of Conceptual Expression. Integration of Morality and Politics: Ancient Chinese humanities emphasize the integration of moral cultivation and political practice, which differs from the Western tradition that leans more towards rationality and scientific research. The All-round Talent View of Literati: The ancient Chinese literati culture emphasizes being well-versed in poetry, literature, and painting, closely linking literary and artistic achievements with personal virtue and social responsibility. Pragmatism Tradition: Chinese humanities historically often emphasize pragmatism, especially knowledge related to governance and statecraft. Overall, the development of Chinese humanities integrates philosophy, literature, art, and moral education, forming a unique cultural tradition and academic system.
Author’s Comment: When questioning the historical discourse and current issues of Chinese humanities, my greatest concern is the potential unfamiliarity, cultural gap, or insufficient emphasis of the large model research team on Chinese history and culture, which is a common problem rooted in the historical inertia of Western centrism since modern times. However, GPT-4’s response seems to be acceptable. In discussing the historical discourse of Chinese humanities, GPT-4 mentions Confucianism, Daoism, the “Six Arts,” poetry, calligraphy, and painting, and also highlights the significance of the imperial examination system. In modern times, it addresses the May Fourth Movement and the influence of Western learning, how modern literature integrates Chinese and Western elements, and how interdisciplinary and global perspectives lead to the Westernization of the academic discourse of Chinese humanities, even touching upon the phenomenon of loss of voice in discourse. All these are within its perspective. GPT-4’s discussion of the historical and contemporary uniqueness of the conceptual expression of Chinese humanities, such as overly moralistic and politicalized wording and the pragmatic expression of humanistic discourse, is indeed one of the cultural characteristics of Chinese humanities. Purely metaphysical or purely aesthetic humanism in the Chinese context is indeed difficult to imagine. Of course, in this historically underdeveloped cultural region where the theoretical advocacy and discourse logic of science and technology have not flourished, the scope of the humanities extends beyond the Western academic categories of linguistics, literary history, and philosophy, covering the entire social sciences. Under the long-term influence of the Confucian cultural concept of “cultivating oneself, regulating the family, governing the country, and establishing peace in the world,” the humanities have always been related to governance and statecraft, making the historical humanities in China far exceed the basic disciplinary scope of Western humanities, becoming a cultural expression of the national ideology centered around Confucian and Daoist classics and historical texts. However, GPT-4 does not mention these. Of course, considering the historical and cultural generation basis and the expectations of the user group for AI applications like ChatGPT, and given that ChatGPT currently possesses only a limited proportion of Chinese cultural corpus data (reportedly less than 5%), it is indeed a “tall order” to make it compensate for and generate related topics in Chinese humanities. Based on GPT-4’s response to this question, it is necessary and important to have it discuss the relationship between the humanities and social sciences.
5. Similarities and Differences Between the Humanities and Social Sciences
Question: Could you discuss the similarities and differences between the humanities and social sciences?
GPT-4: The humanities (Humanities) and social sciences (Social Sciences) are both disciplines that study human behavior and society, but they have significant differences in research methods, research focuses, and theoretical frameworks.
Similarities: 1. Research Object: Both focus on human society, culture, and behavior. 2. Interdisciplinary Nature: Both humanities and social sciences tend to engage in interdisciplinary research, involving a wide range of academic fields and methods. 3. Social Impact: They both contribute to understanding social phenomena, historical development, and cultural differences, having a profound impact on social policy and public life.
Differences: 1. Research Methods: Humanities: Focus more on interpretative and critical analysis, emphasizing text interpretation, historical context analysis, and theoretical critique. Common methods include textual analysis, historical research, and comparative studies. Social Sciences: Tend to use quantitative methods and scientific research methods, such as statistical analysis, experimental design, and survey research. 2. Research Focus: Humanities: Concerned with the uniqueness of individuals and cultures, emphasizing deep interpretations of culture, history, and language. Focus on fields such as art, literature, and philosophy. Social Sciences: Emphasize general laws of social phenomena, focusing on group behavior, social structure, and social relationships, including economics, political science, sociology, etc. 3. Theoretical Framework: Humanities: Explore more abstract concepts and theories, such as moral philosophy, aesthetic theory, and critical theory. Social Sciences: Focus on establishing testable and quantifiable theoretical models, emphasizing the analysis and interpretation of empirical data. 4. Purpose and Application: Humanities: Aim to enhance individuals’ understanding of human experiences and cultural heritage, emphasizing personal self-realization and critical thinking. Social Sciences: Aim to understand and solve social problems, provide policy recommendations, and promote social welfare.
Summary: Although the humanities and social sciences differ in methods and focus, both are indispensable parts of understanding complex human social and cultural phenomena. They complement each other and collectively constitute a comprehensive understanding of human life.
Author’s Comment: In discussing the similarities and differences between the humanities and social sciences, GPT-4 systematically outlines the basic differences in research methods, research focuses, theoretical frameworks, and purposes and applications, but its summary of the similarities is quite inadequate. GPT-4 points out that both the humanities and social sciences “focus on human society, culture, and behavior,” but a more comprehensive and in-depth explanation should be that both study the social, historical, and cultural contexts in which humans are situated, including all social relationships and realities. Furthermore, GPT-4’s summary that interdisciplinary research is a similarity between the humanities and social sciences is indeed a stretch. Within the framework of my question, GPT-4 seems keen to mention “interdisciplinary”; I wonder if this is a so-called “one-size-fits-all” response template. However, it cannot be denied that GPT-4 has logically answered the question and concluded that “both are indispensable parts of understanding complex human social and cultural phenomena,” and “they complement each other and collectively constitute a comprehensive understanding of human life.” This is reasonable and accurately summarizes these two categories of disciplines, especially since current domestic universities and academia often merge the two into “humanities and social sciences.” We can further predict that it will not be long before, as the professional academic applications of ChatGPT become a common cultural ecosystem, each user will effectively have hundreds of academic mentors, research assistants, and intelligent databases at their disposal, potentially leading to exponential growth in research efficiency. This would be particularly beneficial for historical and literary studies in the humanities.
6. The Breakthrough Development of Generative Artificial Intelligence and Its Impact on the Humanities
Question: My question is actually your question. So, can you discuss the recent breakthrough developments in generative artificial intelligence, such as ChatGPT, Gemini, Wenxin Yiyan, etc., and the various applications of these large models in the humanities, such as data collection, translation studies, linguistics, literature, and graphic image video generation apps, and whether they have already had a significant impact on the development of the humanities?
GPT-4: Generative artificial intelligence (AI Generated Content, AIGC), such as ChatGPT, Gemini, Midjourney, Wenxin Yiyan, etc., has indeed begun to have a significant impact on the humanities. The impact of these technologies on the humanities can be viewed from several aspects: 1. Research Methods and Tools, including Data Analysis: AI can process large amounts of text and data, providing new analytical tools for fields such as literature, history, and linguistics. Language Models: AI language models, such as ChatGPT, can help study language use, language development, etc., providing new perspectives for linguistics. Text Generation and Analysis: The application of AI in text generation and analysis provides new tools for literary creation and text criticism. 2. Education and Learning, including Teaching Assistance: AI can serve as a teaching assistant, helping students learn languages, history, philosophy, and other humanities subjects. Customized Learning: AI technology can customize learning content based on students’ learning progress and interests. 3. Creation and Art, including Artistic Creation: In fields such as music, literature, and visual arts, AI has been used to create new works, promoting new ways of creative expression. Literary Works: AI-generated literary works and poetry are exploring new boundaries of language and narrative. 4. Cultural Studies, such as Cultural Analysis: AI can analyze and understand content in different cultural contexts, providing new perspectives for cultural studies. 5. Ethical and Philosophical Discussions, such as AI Ethics: The development of AI has also sparked discussions on machine ethics, human-machine relationships, consciousness, and the responsibilities of artificial intelligence. 6. Paradigm Shifts in Academic Research, such as Interdisciplinary Research: The use of AI promotes the cross-fusion of the humanities with the fields of science and technology. 7. New Problems, Challenges, and Critiques, such as Originality and Copyright Issues: AI-generated content raises discussions about originality and copyright. Quality and Accuracy Issues: The content generated by AI may have varying quality and limited accuracy. The Impact on Human Roles: The role of humans in creation and research raises questions about their value in art and academia.
Overall, generative artificial intelligence provides new tools and methods in the field of humanities, but it also brings a series of ethical, legal, and quality challenges. Its impact on the humanities is complex and multifaceted, requiring ongoing observation and research.
Author’s Comment: I strongly agree with GPT-4’s answer to this question. It almost systematically covers all the impacts that artificial intelligence, especially generative AI represented by ChatGPT since 2022, has had and is having on the humanities, summarizing them comprehensively and systematically, which exceeds even my own understanding as someone who considers themselves knowledgeable about the interdisciplinary research between technology and humanities. When we begin to think about the broad theme of “the iterative development of artificial intelligence and the future development of the humanities,” we realize that merely reading and contemplating in isolation is far from sufficient; rather, relying on the massive training data and the computational power of large-scale servers that support the continuous iteration of artificial intelligence can undoubtedly play a significant role in related research on future trends. In this current era where technology and humanities intertwine and are inseparable, we urgently need to break through the barriers of our own disciplines, embrace artificial intelligence, and learn and explore all aspects of this innovative technology, which may determine the future development of the humanities. Only in this way can we find answers that are closer to the actual status and capability space of future developments in the humanities.
7. Application Examples:
Using ChatGPT and Human Scholars to Study Lu Xun’s Literature
Question: Suppose I want to use generative artificial intelligence to support a literary study on Lu Xun, could you provide some possible research assistance in a concise manner?
GPT-4: Generative artificial intelligence (AI), such as the ChatGPT series of models, can provide numerous assistance and enhancements in the field of Lu Xun’s literary research. Here are some specific applications: 1. Text Analysis and Corpus Construction: AI can quickly and comprehensively analyze all of Lu Xun’s works based on your requirements, extracting themes, styles, narrative techniques, etc. from his complete works, including novels, essays, and papers. Moreover, it can also construct a corpus of Lu Xun’s works and relevant critiques, facilitating in-depth text mining and comparative studies. 2. Style and Theme Exploration: Style Imitation and Analysis: AI models can imitate Lu Xun’s writing style to help researchers understand his unique literary expression. Theme Discovery: AI can identify recurring themes and viewpoints in Lu Xun’s works, revealing the core and evolution of his thoughts. 3. Historical and Cultural Background Research: Background Integration: AI can assist in integrating historical and cultural background materials from Lu Xun’s era, providing a more comprehensive research perspective. Social Impact Analysis: AI can analyze the social impact and cultural significance of Lu Xun’s works on his era and beyond. 4. Cross-Cultural Comparative Research, such as International Perspectives: AI can help compare Lu Xun’s works with those of other countries, promoting cross-cultural understanding. 5. Teaching and Popularization: Teaching Assistance: AI can serve as a teaching tool, helping students understand Lu Xun’s literary works and the historical and cultural background behind them. Popularization: AI-generated content can be used to popularize Lu Xun’s ideas, attracting more interest in his literature. 6. New Perspectives and Innovative Research: New Perspective Development: AI’s deep learning and pattern recognition capabilities may uncover perspectives and connections that traditional research has overlooked. Innovative Research Methods: The use of AI encourages the adoption of more innovative research methods, such as data-driven literary analysis.
Conclusion: Although AI has immense potential in Lu Xun’s research, attention must also be paid to issues of data quality, model bias, and interpretative depth. AI tools should be viewed as aids for human researchers, not as substitutes, and should be combined with traditional literary research methods to ensure depth and quality in research.
Author’s Comment: I asked GPT-4 for “concise prompts,” and its suggestions regarding research methods and academic paper structure are very suitable for novice researchers. As it stands, GPT-4 provides significant assistance in areas such as data collection, background development, research history organization, text analysis, style and theme exploration, and cross-cultural comparison. However, if the research concept and paper planning are more creative and systematically detailed, the support that ChatGPT can provide becomes less predictable. GPT-4 reminds that technically “attention should be paid to data quality, model bias, and interpretative depth issues,” and that “AI tools should be seen as aids for human researchers, not substitutes, and should be combined with traditional literary research methods to ensure depth and quality.” This rational, thorough, and inclusive perspective seems to regard human researchers with admiration, and this AI researcher can indeed be seen as a companion in the study of Lu Xun, rather than an entity to be guarded against, and it may even serve as a machine academic guide.
8. Challenges Posed by Artificial Intelligence to Humanities Research
Question: Could you analyze in detail the negative impacts and potential dangers that the emergence and application of large language models like GPT may bring to humanities research?
GPT-4: The emergence and application of large language models like GPT indeed bring and may bring some negative impacts and potential dangers to humanities research, primarily reflected in the following aspects: 1. Data Quality and Bias. Bias and Misinformation: Since large language models are usually trained on existing internet data, this data may contain biases, misinformation, and outdated views. Such biases may be learned by the model and reflected in its outputs. Content Reliability: The content generated by the model may have accuracy issues, especially when dealing with complex topics in the humanities. 2. Originality and Intellectual Property. Plagiarism and Originality Issues: Models like GPT can generate content similar to existing works, which may raise plagiarism and originality disputes. Copyright Issues: The content generated by the model raises questions about the ownership of intellectual property rights, potentially leading to legal and ethical controversies. 3. Academic Integrity. Dependency Issues: Students and researchers may become overly reliant on AI tools, diminishing their own research and critical thinking abilities. Academic Dishonesty: The use of AI tools may lead to academic misconduct, such as using AI-generated content without proper citation or declaration. 4. Depth and Diversity of Research. Research Depth: Over-reliance on language models may limit researchers’ exploration of themes’ depth and complexity. Intellectual Diversity: AI models may lead to the homogenization of academic thought and research methods. 5. Critical Thinking. Thinking Inertia: Over-reliance on model-generated answers may result in a lack of creativity and critical thinking when solving problems. 6. Professional Ethics and Responsibility. Ethics and Responsibility: Defining the scope of model use, data processing methods, and ethical responsibilities when using AI models for humanities research is a challenge. 7. Human-Machine Relationships. Human Roles: The widespread application of AI in the humanities may lead to a reconsideration of the roles and values of human researchers.
Conclusion: The application of large language models like GPT in humanities research needs to be cautious, and the limitations and potential risks of these tools must be recognized. At the same time, researchers and students should be educated on how to use these tools correctly and maintain critical and ethical awareness when doing so. Ultimately, these tools should be viewed as aids for human researchers, not substitutes.
Author’s Comment: Although GPT-4 is not yet mature, as a significant technological breakthrough approaching AGI, it has already begun to possess various capabilities in data processing, such as seeing, hearing, speaking, reading, writing, calculating, programming, and generating graphical and image language video interactions. This is an unprecedented advancement in human technological innovation history. The once-clear relationships between technology and civilization, technology and humanities, and technology and humans are starting to blur in the face of general artificial intelligence. In terms of AI’s massive parameter training and data storage, the emergence of new narratives and viewpoints that exceed general human intelligence and experiential expectations exists within the black box of unknown potential intelligence capabilities, which ChatGPT itself has already sensed. Of course, this is also a capability bestowed upon it by the creators of models like ChatGPT. At the same time, ChatGPT has made judgments about the disruptive changes that it may bring to various fields of society and has raised warnings through its statements. For some time now, ethical and moral considerations regarding AI have been actively addressed, studied, and attempted to be regulated and managed by governments, educational departments, and academia. Similarly, when generative artificial intelligence inevitably intervenes in the field of humanities research, the limitations and potential risks that accompany it must also become important considerations for academia, and the sooner this is addressed, the better. However, academia should not reject AI outright due to these concerns. Currently, the humanities especially need to transcend their disciplinary barriers and seek new paths for development and enhancement through cross-disciplinary interactions.
9. Predictions from Artificial Intelligence:
Opportunities and Possible Futures for the Humanities Brought by Artificial Intelligence
Question 1: Please analyze the new opportunities and breakthrough cognitions that the rapid iteration of generative artificial intelligence and the possible maturity of AGI may bring to the future development of the humanities.
GPT-4: As generative artificial intelligence (such as GPT) rapidly iterates and artificial general intelligence (AGI) matures, the humanities may experience significant transformations and breakthroughs. Here are some potential new developments and breakthrough cognitions: 1. New Research Tools and Methods. Big Data Analysis: AI can process and analyze vast amounts of text and data, providing new research tools for fields such as linguistics, history, and literature. Pattern Recognition: AI’s ability to recognize complex patterns will help reveal hidden relationships and trends in humanities data. 2. Strengthening Interdisciplinary Research: Fusion of Technology and Humanities: The development of AI will promote the deep integration of technology and humanities, leading to new interdisciplinary research fields. Emergence of New Disciplines: New fields may emerge, such as AI ethics, digital humanities, etc. 3. Accessibility: The proliferation of AI tools will make advanced research methods more accessible, lowering the barriers to academic research. Global Collaboration: AI will facilitate global academic collaboration, making it easier for researchers from different regions and backgrounds to share knowledge and resources. 4. Improved Research Efficiency and Quality. Automated Analysis: AI can automate many research tasks, such as literature reviews, data collection, and preliminary analysis, improving research efficiency. Deep Insights: AI’s advanced analytical capabilities may lead to deeper insights into topics in the humanities. 5. New Teaching Methods. Personalized Learning: AI can provide personalized teaching content and methods, adjusting teaching plans according to students’ learning progress and interests. Interactive Learning Tools: AI can serve as interactive learning tools, enhancing the appeal and effectiveness of education. 6. In-depth Discussions on Ethics and Philosophy. Ethics and Responsibility: The application of AI in the humanities will provoke in-depth discussions about ethics, responsibility, and human roles. Exploration of Cognitive Models: AI’s simulation of human cognitive processes will promote new developments in philosophy and psychology. 7. New Perspectives in Cultural and Social Research. Cultural Pattern Analysis: AI’s pattern recognition capabilities will reveal new patterns in cultural phenomena and social changes. Real-time Analysis of Social Dynamics: AI will enable real-time analysis and understanding of social dynamics and cultural trends.
Conclusion: Although the development of generative artificial intelligence and AGI presents tremendous potential for the humanities, it also comes with challenges, particularly concerning ethics, bias, and data quality. In the future, research in the humanities will need to effectively combine these new technologies with traditional research methods while addressing their complexities and potential risks.
Question 2: Could you provide a more specific and detailed prediction of how the various branches of the humanities may evolve in 2035, after more than a decade of iteration of generative artificial intelligence?
GPT-4: Predicting the potential impacts of generative artificial intelligence on various branches of the humanities in the next decade requires consideration of the trends in AI technology and the characteristics of the humanities. Here are some reasonable predictions: 1. Linguistics. Breakthroughs in Natural Language Processing: AI technology will further enhance its ability to process natural language, particularly in understanding and generating various languages and dialects. Language Change Research: AI will be able to more effectively track and analyze the evolution of language, including the development of dialects and slang. 2. Literature. Deepening Text Analysis: AI will be able to analyze text structures, styles, and themes more deeply, even simulating the writing styles of different authors. Innovations in Creative Tools: AI will be widely used as a new creative tool, helping writers and poets explore new literary forms and narrative methods. 3. History. Reconstruction and Simulation of Historical Scenes: AI will enable more accurate reconstruction and simulation of historical events and social changes, providing visualizations of different historical hypotheses. Digital Archives: AI will assist in constructing more comprehensive digital archives, making the retrieval and analysis of historical documents more efficient. 4. Philosophy. Exploration of Ethical and Philosophical Issues: With the development of AI, new philosophical and ethical questions will arise, such as consciousness, free will, and moral responsibility of artificial intelligence. Logic and Thought Experiments: AI will be used for complex logical analysis and thought experiments, helping philosophers explore intricate philosophical theories. 5. Art History. Analysis of Artworks: AI will be able to conduct deeper visual analyses of artworks, identifying styles, techniques, and historical contexts. Predicting Trends in Art Development: AI may be used to forecast the development trends of art styles and movements. 6. Religious Studies. Analysis of Religious Texts: AI will conduct deeper analyses of religious texts, exploring the interrelationships between different religious traditions and belief systems. Research on Faith Patterns and Dissemination: AI will analyze the evolution of faith patterns and the dissemination of religious concepts. 7. Cultural Studies. Cross-Cultural Comparative Research: AI will facilitate deeper cross-cultural comparative studies, understanding the similarities and differences between different cultures. Social Dynamics Analysis: AI will be able to analyze social dynamics and cultural trends in real-time, providing in-depth insights into social changes.
Summary: In the next decade, the development of generative artificial intelligence may push the humanities into a new stage, particularly in the enhancement of data processing and analytical capabilities, which will bring revolutionary changes to humanities research. However, this technological advancement also comes with ethical and moral challenges, especially regarding the objectivity and depth of research. The future of the humanities will increasingly depend on the effective combination of artificial intelligence and traditional research methods.
Author’s Comment: The above two questions encompass a general judgment on the potential future development of the humanities in the context of the rapidly evolving environment of artificial intelligence, and a prediction of how various major branches of the humanities may evolve around 2035. As a premise, GPT-4 emphasizes that this all depends on the rate of AI iteration and the technological heights that general artificial intelligence can achieve. Nevertheless, regardless of the current pace of progress, the impact of AI on the humanities as a whole and on various branches will undoubtedly be immense, indicating that “the humanities may experience significant transformations and breakthroughs,” such as “revealing hidden relationships and trends in humanities data,” “promoting the deep integration of technology and humanities,” and creating “entirely new fields of study, such as AI ethics and digital humanities.” At the same time, AI will undoubtedly have the capacity to automate many academic research tasks, such as “literature reviews, data collection, and preliminary analysis,” leading to a deeper understanding of themes in the humanities. Furthermore, based on a strong grasp of data regarding authors’ creations and their environments, we may be able to visually reconstruct historical scenes of literary creation. AI will likely enable high-fidelity simulated reconstructions and research of historical, literary, artistic, and even philosophical facts, giving rise to entirely new intelligent humanities. Humans may even attempt to conduct visualized, scenarized, and process-oriented logical and thought experiments. Of course, machine writing will likely be able to precisely simulate any renowned author’s writing style, potentially attempting new literary types and writing methods, producing new genre-based machine literature that resonates with readers.
In short, future advancements in AI iteration and disruptive technological breakthroughs will likely “push the humanities into a new phase.” Even the current explosive progress of artificial intelligence presents more opportunities for the humanities to break away from the existing rigid hierarchical and marginalized structures of disciplines in science, technology, medicine, and business. With the assistance of AI’s “intelligent fast track,” the humanities may be able to shed the criticisms of being unscientific, non-technical, and non-quantitative, thereby unleashing a tremendous cultural productivity in critical thinking, innovative thought, and knowledge construction, becoming a new discipline in the intelligent era. I have always believed that by observing the historical relationship between technology and civilization, one can make a basic judgment: technology is the foundational driving force behind the development of civilization and culture, and it is also the inherent driving force behind the development and renewal of the humanities. Individual human researchers may lose their guardianship and control over data information, academic history, and disciplinary history in the age of artificial intelligence, no longer being the custodians of a vast sea of human academic inventory, but they may become stronger innovators in the humanities and pioneering thinkers of original ideas.
10. Conclusion
Today, human society has reached an important historical juncture. As a summary of the above dialogue, my overall judgment regarding the potential future development of artificial intelligence and the humanities is that, although it is currently difficult to accurately predict the breakthrough development opportunities and impacts that rapidly evolving artificial intelligence, especially generative artificial intelligence or AGI, may bring to human social production and life, we are on the eve of a significant transformation in the humanities and many other disciplines, even the entire academic system, in the face of the current state and foreseeable future of new artificial intelligence, quantum science, biological science, material science, and other major disruptive breakthroughs. Just observing the societal response to ChatGPT in 2023 and the millions of astonishing application apps that appeared within a year of the launch of OpenAI’s store, one can foresee the immense influence of artificial intelligence. Specifically regarding the development of the humanities, GPT-4 has repeatedly emphasized that research in the humanities needs to utilize new technologies in artificial intelligence while also addressing the complexities and potential risks they bring.
First, artificial intelligence is likely to trigger an overall change in our concept of the humanities disciplines. From the writing patterns and structural processes of this article, it can be seen that artificial intelligence is already participating in the research and formation of papers as a relatively autonomous dialogue author and text creator. As a so-called “another author,” artificial intelligence has significantly changed the cultural production subjects, production methods, and results presentation patterns in traditional literary research. The new humanities production subjects brought by artificial intelligence (as a non-biological author), new humanities production methods (human-machine interactive dialogue and machine intelligence generation), and new forms of humanities results, namely the formation and presentation patterns of texts (non-temporal writing, instant text presentation, etc.), signify that a new research paradigm and disciplinary structure with ontological significance is beginning to form. Some new fields, such as computational humanities, humanistic computing, process ethics research, data value sociology theory, machine writing theory research, and modeling and simulation of historical events, will gradually become important components of the humanities. Seizing this opportunity, researchers in the humanities will have numerous possibilities to expand the depth and breadth of their disciplines, which will become an indispensable internal demand in this intelligent era.
Secondly, the research paradigms and methods of the humanities are undergoing important changes. For example, previously, almost all humanities research maintained a loose relationship with natural sciences and technological sciences, but entering the era of artificial intelligence, the situation is entirely different. Artificial intelligence, as a form of “machine intelligence,” has, to some extent, already transcended the traditional passive tool attributes of scientific and technological disciplines, becoming a new technology with a certain degree of proactivity and autonomy. The anthropomorphic characteristics formed under human design have endowed artificial intelligence with a certain “autonomy” in humanistic production, producing many unexpected new humanistic products, such as machine-translated works, machine poetry, machine novels, machine dramas, and machine literary critiques, etc. This means that the literary and cultural creation involving artificial intelligence is no longer simply controlled and comprehensively grasped by humans from the outset, but inevitably gives rise to many new propositions and meanings that transcend the interface between technology and humanities. This is a situation that has not been anticipated by the scientific and technological community since the Industrial Revolution, and for the humanities, which aim to study such phenomena, it is almost impossible to conduct in-depth research without employing integrated interdisciplinary methods. This may partially explain why GPT-4 frequently mentions interdisciplinarity. At the same time, breakthroughs and innovations in academic methodologies will lead to a profound elevation of ontological cognition regarding the humanities, which is also worthy of attention and anticipation.
Of course, given the bottlenecks encountered in the current progress of artificial intelligence technology, ChatGPT still faces many technical thresholds it cannot yet cross, such as the influence of training on false data, which may produce almost machine hallucination-like false texts. Without the knowledge judgment and correction from human scholars, such texts are likely to mislead readers. ChatGPT lacks self-awareness; it is merely a passive participant in humanistic academic research. Its language generation method (machine inference logic) fundamentally differs from the human language generation method (consciousness-based language presentation), so it cannot self-reflect or bear responsibility. However, it is undeniable that the disruptive development of artificial intelligence possesses immense potential energy that will bring immeasurable possibilities. Nevertheless, any progress or breakthrough in AI is ultimately the result of the developers and controllers of the technology, and while artificial intelligence has not yet escaped the “如来佛” (Tathāgata) of humanity, it is challenging to predict how general artificial intelligence will develop in the future, which is a topic of ongoing discussion and caution among governments, leading companies, and top AI experts worldwide.
As humanists, we certainly cannot take this lightly, but this is also not a reason for the humanities to reject the future world of artificial intelligence and to advance towards a new stage of development with its assistance, cooperation, and co-creation. Humanists need to remind themselves that entering the rapidly changing era of artificial intelligence requires breaking away from a closed sense of disciplinary self-sufficiency, eliminating the psychological rejection of technology, and establishing an awareness of attention and learning towards technology. We must maintain a scholarly inquiry mindset that remains vigilant about the increasing influence of artificial intelligence. We have witnessed the rapid development of generative artificial intelligence. In January 2024, the AI Store’s 3 million applications have just been launched, and an application reportedly in a hundred billion dollar market is about to expand globally. New breakthroughs seem to be on the horizon. The humanities, along with almost all other disciplines, will likely find it challenging to respond calmly in this context. In the interactive relationship between technology and civilization, just as the economic base determines the superstructure, humanists cannot change this relationship model but must adapt to its rules. Technological advancements will drive the iterative upgrade of the humanities, which seems to align with the historical patterns of humanistic evolution since the inception of civilization.
Finally, let us imagine the future AI humanities world with a bit of creativity. Currently, everyone interacts with GPT via computers, tablets, or smartphones, typing away on keyboards to unfold our paper writing. However, looking at today’s technological development, in the foreseeable future, these tools may be replaced by intelligent glasses, flexible display skins, virtual holographic screens, and other media. We may even likely have a humanoid digital robot that can be summoned and dismissed at will, serving as our work assistant and discussion partner at any time and place. The once research spaces such as studies, libraries, and studios may no longer just be filled with countless books but could evolve into a metaverse-style, fully simulated, fully interactive, and interconnected spiritual world that spans the entire historical tunnel and cultural space of humanity. Humanistic creation and research may gradually integrate into one, becoming a free kingdom of scholarship. If one were in the 18th century, such imaginings would be difficult even for science fiction writers, but in today’s rapidly developing era of artificial intelligence, quantum science, synthetic biology, intelligent material manufacturing, and more, it is not far-fetched to say that this could be the reality in 10 or 20 years. I maintain a cautiously optimistic attitude and expectation regarding this possibility.
(This article is a research outcome from the “Guangdong-Hong Kong-Macao Greater Bay Area Science and Humanities and Innovative Cultural Research Center,” a key research base for humanities and social sciences in Shenzhen.)
END
For article reprints, submissions, and collaborations, please email
