In just two months, ChatGPT has amassed over 100 million monthly active users, becoming the fastest-growing consumer application in history. Developed by OpenAI, this consumer-grade artificial intelligence application is a chatbot that can complete tasks such as writing emails, code, copy, homework, and translations.
While the introduction of ChatGPT has brought great excitement to the rapidly declining global information market, it has also brought anxiety and unease. As AI rises, what challenges does it bring to the grand narrative of the future? The tech industry has begun a new AI arms race, with Gmail's founder claiming that ChatGPT will destroy Google. Google has quickly invested in a ChatGPT competitor (whose founder is also from OpenAI), and many companies have announced upcoming ChatGPT-like projects.
As a consumer-grade application, ChatGPT has brought both surprise and anxiety to the masses. Its impressive capabilities have many questioning whether their jobs are at risk. Will engineers, news editors, translators, market analysts, and customer service representatives be replaced by ChatGPT? Will we become "useless" in our lifetime? Strong AI will raise education costs and reduce birth rates, eventually replacing humans with AI "descendants" in our consciousness.
This article explores the economic impact of strong artificial intelligence from the perspective of competition and elimination.
Artificial Intelligence and Security Boundaries
Many technology enthusiasts have successfully used ChatGPT through various methods. Based on their test feedback, ChatGPT may be the first consumer-level application of strong artificial intelligence. Compared to AlphaGo, ChatGPT has a wider range of consumer applications. Compared to Siri and Microsoft XiaoBing, ChatGPT is based on the GPT model, which is more powerful, has stronger natural language generation capabilities, higher computing efficiency, and can also self-train based on historical conversations.
Currently, what level is ChatGPT at? Elementary school, middle school, or undergraduate level?
Excitingly, ChatGPT is not just a rote learning machine. It can understand the "implications" in the context of language. However, it is slightly disappointing and tiring that ChatGPT is enthusiastic about "meaningless literature". Its answers are logically impeccable, but it seems to say nothing. Its answers to political questions follow a similar pattern to the American politicians, and its analysis of monetary policy follows a pattern similar to former Federal Reserve Chairman Greenspan.
For example, one tester asked: What impact would the US economy and inflation have if the Federal Reserve continues to raise interest rates by 25 basis points in March?
ChatGPT is capable of providing comprehensive analysis frameworks for many questions with grammatically correct and logically complete answers but lacks specific and detailed analyses. The answers provided are at the level of undergraduate knowledge, suitable for establishing basic frameworks for analyzing problems, but may not be able to solve specific, real-time, and complex issues. Currently, the most successful application of ChatGPT is by American college students using it to do homework, with many even achieving the top grade in the class. According to a survey, 89% of American college students admit to using ChatGPT for homework and 53% for writing papers. If you are a college student, you can use ChatGPT to help build your basic logic and general framework for analyzing many problems.
But what about teachers? A student at Northern Michigan University used ChatGPT to generate a philosophy essay and received the highest grade in the class. The professor couldn't tell if it was written by the student or the AI. For mediocre professors who only provide vague and unhelpful feedback, ChatGPT could be the terminator. No professor can match the comprehensive analysis and general framework offered by ChatGPT.
In the future, strong AI will provide correct nonsense and solve specific problems. Through deep learning and self-training, strong AI can achieve a level of sophistication that can replace many specialized jobs. After two months of testing, ChatGPT's answers to the same question are rapidly improving. It will continually self-correct and provide more professional answers based on historical conversations and the latest data.
As one of the founders of OpenAI, Musk praised ChatGPT, saying, "It's scary good. We're not far from dangerous and powerful AI."
I think there are two points worth paying attention to regarding the potential challenges that ChatGPT may bring:
First, the power of artificial intelligence is not in "intelligence," but in computation. Jobs based on computation are easy to replace with AI.
The working principle of AI is based on supercomputing power and data feeding. ChatGPT's advantage lies in its model and algorithm, while Microsoft Xiaobing's advantage lies in its database. Microsoft has announced the integration of many of its products, including Bing and Office, into ChatGPT. If ChatGPT receives higher data support, its computing power will increasingly surpass that of humans in many fields.
In fact, vertically strong AI is more likely to obtain commercial value, such as code robots, companion robots, and AI legal advisers. In the United States, the implementation of precedent law means that judges largely rely on historical precedents to make judgments. However, currently, this system faces the challenge that as the number of precedents in maritime law countries increases over time, today's judges and lawyers cannot read through the vast sea of historical precedents in their lifetime. Professional AI legal advisers can solve this problem, and robots can search and read all precedents and legal texts in a short period of time. As long as the model is good enough, robots can also consider and compare the process, scene, and historical background of each precedent. Although robots are not perfect, they are powerful in calculating case files compared to human beings. Non-licensed legal advisers are not competitive compared to AI legal advisers.
In front of professional AI, jobs such as financial data analysis, news editing, code writing, image generation, music creation, customer service, chat companionship, and legal consultation can all be completed through computation. However, creative work, such as initial program design architecture, product development, business decision-making, mental healing, diplomatic strategy, and judge's verdicts, is not easily replaced at present.
Second, ChatGPT poses a greater threat to general cognitive workers, while relatively complex manual laborers are currently safe.
Many knowledge workers have studied hard for twenty years to improve their knowledge, but they may be replaced by artificial intelligence overnight. The reason is that many people, only memorize and remember knowledge, and master mechanical problem-solving routines and analytical frameworks. These skills can all be defined as storage and computation, the latter being the unparalleled professional ability of artificial intelligence.
They mistakenly believe knowledge can construct a secure boundary of competition, but ChatGPTs quickly defeat them. Strong AI represented by ChatGPT poses a challenge to knowledge workers such as low- and mid-level lawyers, teachers, designers, analysts, engineers, editors, and customer service personnel. For example, when an event quickly spreads on social media, AI editor algorithms can quickly capture, integrate, and edit information to produce a news story. Code robots can quickly write primary code using a vast amount of open-source code.
Conversely, besides manual labor, such as cashiers, gas station attendants, and ticket sellers, physical labor such as playing sports, hairdressing, and massage is not easily replaceable at present. For robots, the difficulty of manual labor is much higher than that of cognitive work. Mechanical operation also depends on hardware revolution and technological progress in other fields, and it is currently not easy to solve the mechanical stability problem of robots.
Can we still keep our jobs?
ILLA Cloud
ILLA Cloud is a low-code development platform with dozens of front-end components and database API integrations. You can use ILLA Cloud to build the front-end interface by dragging and dropping components and connecting to your database or API to complete full-stack development quickly.
ILLA proudly announces a partnership with Hugging Face, a suite of natural language processing (NLP) tools and services. They are most well-known for their open-source NLP library, which provides text generation, language translation, and named entity recognition tools. With Hugging Face, ILLA is more productive than before. Our users can do more with AI.
ILLA Cloud provides dozens of commonly used front-end components, allowing you to build different front-end interfaces based on your specific needs quickly. At the same time, ILLA offers a connection to Hugging Face, allowing you to quickly connect to the API, send requests, and receive returned data. By connecting the API and front-end components, you can implement the requirement that users can enter content through the front end and submit it to the API. The API returns the generated content to be displayed on the front end.
You can check ILLA’s website here at: illacloud.com
GitHub page: github.com/illacloud/illa-builder
Join Discord community: discord.com/invite/illacloud