Why ChatGPT Sparks Panic and What Its Real Technical Foundations Are

Why ChatGPT’s emergence causes panic?

Wu Jun notes that while ChatGPT is a hot topic in China, it has largely faded from discussion in the United States, illustrating a pattern where new technologies receive more hype in Chinese media than in their country of origin.

He argues that excessive hype leads to opportunists profiting from trends, citing past examples such as blockchain and the metaverse, and warns that similar speculation surrounds ChatGPT.

What is the technical basis of ChatGPT?

ChatGPT relies on a language model, a mathematical model first developed in 1972 by Fred Jelinek’s team at IBM for speech recognition, later applied to machine translation and question answering.

The model generates text by selecting the most probable sequences based on massive computation and data; modern versions use billions of parameters and require extensive GPU resources.

Training such models also involves human reviewers who evaluate generated outputs.

How did early language models work?

Early models in the 1990s used simple statistical methods and were inaccurate, prompting the inclusion of syntax, topics, and semantics, which increased complexity and computational demands.

Wu Jun’s own 2000‑parameter model required 20 servers and three months of training; today’s ChatGPT uses around 200 billion parameters.

What kinds of questions can computers answer?

Simple factual questions have clear answers, while complex questions require integrating information, a capability demonstrated by Google’s 2014 QA system and now by ChatGPT.

Can computers write better than humans?

Wu Jun shows examples of AI‑generated poetry based on patterns from classical Chinese poems, explaining that the process involves segmenting text and recombining it according to probability.

He emphasizes that such output is not mysterious; it works well for fixed formats like poems or reports.

What impact does ChatGPT have?

The technology is not a revolutionary breakthrough but an extension of existing language models powered by deep learning and massive compute.

It mainly threatens jobs that involve repetitive content generation, while creators of original knowledge remain indispensable.

What new opportunities does ChatGPT create?

Because of its high resource cost, only those who provide compute resources or related services benefit financially; most users should focus on their core work rather than chasing speculative opportunities.

Key takeaways

Do not fear ChatGPT; understand its scientific principles, avoid chasing hype, and recognize that genuine knowledge creation cannot be replaced by large language models.