DeepSeek inspires the 'deep seek' for humanity – Human-centered AI

Editor's note: With the launch of DeepSeek's open-source AI model, the barriers to entry and the costs of AI development have significantly decreased. It is foreseeable that AI's role in human society will continue to expand, both in its impact on the labor market and its integration into daily life. As the annual Two Sessions commences, CGTN's Sci-Tech Section focuses on key issues surrounding artificial intelligence and launches the "DeepSeek inspires the 'deep seek' for humanity" series. Authored by Li Mingling, an assistant researcher at the Institute of Public Policy at South China University of Technology, these articles explore some critical questions, including how we can ensure that technological progress – while promoting fairness, justice, and human welfare – does not diminish human value. The article reflects the author's opinions, and not necessarily those of CGTN.

Artificial intelligence (AI) may trigger the Fourth Industrial Revolution. As AI technology continues to mature, its widespread adoption raises important questions: How will AI be utilized to benefit people? What impact will it have on the job market and employment structure? These questions, of course, assume that AI is fundamentally a tool.

When I asked DeepSeek what it can do for human beings, it gave the following reply (simplified):

a. Knowledge Integration and Learning Assistance: Simplification of complex information, personalized learning support, and cross-disciplinary knowledge connections.

b. Efficiency Enhancement and Decision Support: Automation of repetitive tasks, multi-dimensional analysis, and real-time information integration.

c. Intelligent Support for Daily Life Scenarios: Management of daily tasks, emotional companionship and communication, cultural dissemination, and education.

d. Creative Inspiration and Collaborative Innovation: Breaking conventional thinking patterns, enhancing collaborative efficiency, and promoting equitable access to technology.

e. Balancing Ethics and Responsibility: Risk alerts, value alignment, and sustainability recommendations.

Many users have highlighted DeepSeek's exceptional performance on social media, showcasing its ability to assist with tasks such as designing travel plans, proposing judicial responses, and developing hotel pricing strategies, among others. DeepSeek has gained significant popularity, demonstrating its ability to meet people's real needs in specific scenarios. Moreover, it combines personalized insights to deliver tailored responses, rather than relying on generic, formulaic content.

Some see AI as an auxiliary tool, or even a substitute, for handling specific tasks. On one hand, AI is designed for practical utility, helping us achieve specific goals and ensuring a proper means-end relationship. On the other hand, AI leverages big data and deep learning to achieve intelligence and automation, surpassing the abilities of ordinary humans. This, in turn, raises an important question: Should tools serve people, or replace them?

We must also acknowledge that the instrumental nature of AI has two inherent shortcomings.

First, AI's data sources are limited; it cannot explore things that have not been digitized, and it struggles to generate higher-order cognitive insights.

Second, AI's detachment from human perception prevents it from replacing the genuine experience of the human body, emotions, and actions.

These two limitations underscore the importance of human innovation and practical experience. This suggests that, in the future, humans should place greater emphasis on non-digitized tacit and practical knowledge, such as local informal norms, empirical research, and similar forms of expertise.

In her book "The Worlds I See," AI scientist Fei-Fei Li advocates that while AI technology has the potential for great transformative change, its ultimate goal is to enhance human well-being and quality of life. In 2019, she co-founded the Stanford Institute for Human-Centered Artificial Intelligence (HAI). In its introduction, HAI outlines the rationale behind its establishment:

Humanity faces radical historical changes, and to make these changes beneficial, we need to make AI capable of understanding human beings. For this, the creators of AI must represent different social groups and a diverse array of disciplines. The collective responsibility of AI creators is to guide AI to positively affect the planet, nations, communities, families and people's lives.

HAI embodies both the vigilance of AI scientists and their reverence for human society, reflecting a certain level of ethical awareness and social responsibility. Academic researchers often refer to Human-Centered Artificial Intelligence as HCAI for short.

Research on HCAI has surged in the past two years. Among those contributing to the field, Rezaev, a professor at Rikkyo University in Japan, cautions against turning human-centered AI into merely another buzzword. He stresses the need to clearly define what it truly means for AI to be "good for humans."

This brings us to the first core principle of HCAI: AI should serve the common interests of humanity, not replace or control us.

Health, security, peace, and happiness represent the common interests of humanity. In pursuit of these goals, scientists have sought to apply AI to various fields such as medical diagnosis, environmental and climate monitoring, elderly care, digital governance, and education. They aim to extend these technologies to grassroots communities, underserved areas, and low-income groups, with the goal of overcoming the social and structural barriers that limit access to essential public services.

However, how can we minimize the negative impacts of AI technology?

This leads to the second key principle of HCAI: mitigating the secondary risks that AI may pose and ensuring that AI becomes a symbiotic partner in the development of human civilization.

Of course, this aligns with the vision of building a community with a shared future for humankind. Only through continuous global dialogue, consultation, and common governance can we effectively address the challenges posed by AI.

In short, in an era of AI rich in humanity, technology and society are in harmony, and instrumental rationality and value rationality are unified.

For more:

DeepSeek inspires the 'deep seek' for humanity – Democratization of AI

DeepSeek inspires the 'deep seek' for humanity – Socialization of AI

(Cover: VCG)