The Pros and Cons of Artificial Intelligence: From the Dartmouth Conference to the Joint Boycott

In October 2025, the technology industry sparked a profound discussion on the future development of artificial intelligence (AI). Over 800 prominent figures jointly issued an open statement, calling for an immediate pause in the research and development of "superintelligence". Among the signatories were representatives from various fields, including Steve Wozniak, co-founder of Apple, and Yoshua Bengio, the "father of modern AI". Behind this appeal lies their deep-seated concern about the potential risks of superintelligence—ranging from "the loss of human economic value and deprivation of power, the collapse of freedom and civil rights, the disintegration of dignity and control, to national security risks and even the potential extinction of humanity". This incident has triggered a global re-examination and reflection on the pros and cons of AI.

The origin of AI can be traced back to the 1956 Dartmouth Conference, where John McCarthy first proposed the concept of "artificial intelligence". The goal of this emerging discipline is to simulate, extend, and expand human intelligence. In the early days, AI was full of promise. In 1959, a checkers program with learning capabilities was already able to defeat its designer, and three years later, it beat the checkers champion of a U.S. state. These early successes gave rise to excessive optimism; researchers attempted to tackle more complex tasks but encountered one failure after another, foreshadowing that the development path of AI would be fraught with twists and turns.

The development of AI has gone through numerous peaks and troughs. In the 1970s, research on AI algorithms fell into a slump, while hardware and systems flourished. By the 1980s, the emergence of expert systems transformed AI from theoretical research to practical application, achieving success in fields such as medicine, chemistry, and geology. However, as the scale of application expanded, the limitations of expert systems gradually became apparent—narrow application scope, lack of common-sense knowledge, and difficulties in knowledge acquisition. It was not until 1997, when IBM's "Deep Blue" computer defeated Garry Kasparov, the world chess champion, that another upsurge in AI was ignited. This breakthrough marked the beginning of AI technology moving toward a new height.

Entering the 21st century, with the development of information technologies such as big data, cloud computing, the Internet, and the Internet of Things, AI finally ushered in a new upsurge of explosive growth. In 2016, AlphaGo, developed by Google's DeepMind, defeated the world Go champion, sparking a boom in deep learning across various application scenarios, including natural language processing. Subsequently, OpenAI released models such as GPT-3, ChatGPT, and Sora, which demonstrated powerful capabilities in text generation and video generation respectively, opening the prelude to large language models. AI truly achieved a technological breakthrough from "unusable and inconvenient to use" to "usable", penetrating into all aspects of our lives. The technological progress during this period has expanded the application scope of AI to an unprecedented extent.

The benefits brought by AI are obvious. It can automatically process large amounts of data and perform repetitive tasks, significantly improving work efficiency. In the manufacturing industry, AI can optimize production lines and processes, enhance production efficiency, and reduce labor costs. In the medical field, AI can assist doctors in quickly diagnosing diseases and formulating treatment plans, reducing the possibility of misdiagnosis and missed diagnosis. From intelligent transportation to smart homes, and from financial services to e-commerce, AI technology is bringing innovations to various fields and providing users with more convenient and personalized services. These practical applications fully demonstrate the great potential of AI technology.

However, the development of AI is also accompanied by many risks and challenges. With the popularization of AI technology, the risk of social unemployment is increasing. Data privacy and security issues also cannot be ignored. AI technology needs to process a large amount of data; if data protection measures are not in place, users' sensitive information may be leaked. More worrying is that the use of AI may lead to the degradation of human beings' own abilities. An experiment conducted by the MIT Media Lab showed that the brain activity of people who use AI for writing is significantly reduced. A study in Switzerland also found that people who frequently use AI tools have significantly lower scores in critical thinking. These issues all require us to take them seriously and conduct in-depth thinking.

Currently, AI has triggered a series of moral and ethical issues, such as whether machines have self-awareness and whether they should have rights. The development prospect of superintelligence has even aroused widespread concern. The joint statement signed by over 800 prominent figures warns that superintelligence may bring risks ranging from the loss of human economic value and deprivation of power to the potential extinction of humanity. These concerns have led to a rare consensus in the technology industry—people from different backgrounds have jointly expressed their concerns about the development direction of AI. This phenomenon reflects that the development of AI has entered a new stage that requires careful consideration.

Faced with the pros and cons of AI, we need to make wise choices. In fact, AI will not automatically make people dumber or smarter; what really matters is how people use it. If we treat AI as a crutch and rely on it for everything, our brains will gradually "rust". But if we regard AI as a tool, letting it handle tedious tasks and focusing our energy on thinking and creation, AI can become an extension of our capabilities. This choice lies in the hands of each individual. Today, AI has just become popular, and many usage habits are still being formed. The habits we develop now may affect the development direction of AI in the next few decades.

The development of AI is at a crossroads. On one side are the warnings from AI pioneers, and on the other side are the continuous advances of technology companies in pushing the boundaries of innovation. As the signatories of the joint statement have warned, before reaching a strong public consensus and broad scientific agreement, and confirming that its development is safe and controllable, we may really need to pause and think about where the path of superintelligence will lead humanity. Mankind not only needs to create intelligence but also needs to control the direction of wisdom. At this critical historical moment, we need to face the opportunities and challenges brought by AI with a more prudent and responsible attitude, ensuring that this powerful technology can truly benefit human society.