ChatGPT is a user-friendly chat bot currently accessible to the public. It is the hot new product of OpenAI, a San Francisco-based firm whose founders need no introduction in Silicon Valley circles (Elon Musk and Peter Thiel among others). Current CEO Samuel Altman previously did work for Dropbox, Airbnb and Reddit.
The chatbot was launched two months ago in a free trial format. Already it has gained millions of followers and has helped craft stories for the Atlantic and New York Times, if only to offer a dystopian taste of what’s to come. Chatbots have been around for a while, but this one is known for its effectiveness in generating human-like prose and poetry that’s almost indistinguishable from the real thing.
When I queried the chatbot about China-US Focus, it produced this typed response in under a minute:
"China-US Focus is an online news and analysis website that focuses on the relationship between China and the United States. The website aims to provide objective and accurate information and analysis on issues related to the relationship between the two countries, including economic, political, and strategic matters.
The website is owned by China International Publishing Group (CIPG), which is a state-owned publisher in China. The website is also run by a group of Chinese scholars, analysts, and experts on China-US relations. The website's articles are written by a mix of Chinese and American authors who are experts in their fields.”
Allowing for some problems with repetition, tone and minor inaccuracies, it sure sounds convincing. If it lives up to the hype, it will be a game-changer for Internet search, education, research and writing.
The product is a bit akin to computer-generated images in the sense that it provokes the uncanny valley of perception, hovering somewhere between the real and unreal. Despite its appealing ability to mimic reality, it’s often “off,” if only slightly.
On the other hand, the very premise of AI is to create systems that are self-taught to get better all the time, and this bot, or its progeny are sure to be even more awesome, and more troubling, too.
The chatbot will soon be monetized, its business model predicated on its appeal to the corporate world where it will carry out tasks and save money in the long run by reducing the need for...people.
If robotics has dealt a body blow to certain blue-collar professions, ChatGPT threatens to do the same for many white-collar jobs that require producing reports and data handling. Judge for yourself if contributors to China-US Focus face potential competition from this hot new bot.
When asked the thorny but important question about how U.S.-China relations might be improved, Open AI’s Chat Generative Pre-trained Transformer (GPT) wrote out these bullet points in less than a minute:
- Diplomatic engagement: Both countries could engage in more frequent and high-level diplomatic talks to address areas of tension and find common ground.
- Economic cooperation: The two countries could seek to cooperate on economic issues, such as trade and investment, to reduce tensions and promote mutual growth.
- Managing territorial disputes: The two sides can work to manage and reduce tensions around territorial disputes in the South China Sea and Taiwan.
- Human rights: The US and China could engage in dialogue on human rights issues, including those related to Xinjiang and Hong Kong, in order to find a way to address these concerns.
- Military: The two countries could engage in confidence-building measures and reduce the risk of accidental military clashes
- Transparency and communication: Both sides could increase transparency and communication to reduce misunderstandings and the risk of miscalculation.
When given a prompt about U.S. politicians who are anti-China, it produced a short list:
- Senator Tom Cotton of Arkansas, who has been a vocal critic of China's economic and military policies, as well as its human rights record.
- Senator Marsha Blackburn of Tennessee, who has also been critical of China's economic and military policies and has advocated for a more aggressive approach to dealing with China.
- Secretary of State Mike Pompeo, who served as the Director of the CIA before becoming the Secretary of State, was known for his critical stance on China. He has been a vocal critic of China's human rights record and has taken a hardline stance on US-China relations during his tenure.
- Senator Marco Rubio of Florida, who has been critical of China's human rights record and has advocated for a more aggressive approach to dealing with China.
- Senator Mitt Romney of Utah, who has also been critical of China's human rights record
When asked about voices pro and con China in the journalist community, it hedged a bit before offering a pair of examples from each side:
“It is difficult to provide a comprehensive list of US journalists who are pro-China or critical of China, as views on China can vary among journalists, and can change over time. However, I can provide some examples of journalists who have been known for their coverage of China and their views on US-China relations:
- Susan Shirk, a former Deputy Assistant Secretary of State for East Asian and PacificAffairs, who has written extensively on US-China relations and has advocated for a more cooperative approach to dealing with China.
- Orville Schell, an American journalist, author and academic who has written extensively on US-China relations and has advocated for a more cooperative approach to dealing with China.
- Gordon G. Chang, an American journalist, author, and blogger, who has written extensively on US-China relations and has been critical of China's human rights record and its economic and military policies.
- Michael Pillsbury, an American journalist, author and China expert, who has written extensively on US-China relations and has been critical of China's human rights record, its economic and military policies, and its role in the world.
OpenAI is trying to avoid legal tangles by “taming” the mind of the computer beast. There are areas where this is arguably a social good, especially concerning hate sites, racist rants, antisemitism and so on, but it raises the question about where the mother ship draws the line for its clever globe-roving chatbot.
To date, OpenAI has been opaque about the way it filters information. Its chat bot won’t answer questions about its “creator” (feigning ignorance) but the bot must filter to meet market needs, both in the U.S. and elsewhere. Curious about how it filters itself, I asked ChatGPT about how it deals with China, given China’s own unique notions of free speech, protected speech and hate speech.
“OpenAI, like any other company operating in China, is subject to Chinese laws and regulations, and it may need to cooperate with the Chinese government in order to comply with them…OpenAI has not made any public statements about its cooperation with the Chinese government, therefore I can't confirm that.”
It is anticipated that Chinese search engine Baidu will launch an automated chatbot service similar to ChatGPT in the spring, though details remain scant at this point. Whether China will embrace ChatGPT or come up with its own robust alternative remains to be seen, but the cat’s out of the bag now and developments in AI will be closely watched.
To put things in perspective, I’ll let the chatbot get the last word on what it is, and what it isn’t.
“I am a language model and I don't have consciousness, emotions or personal experiences. My responses are based on patterns and information found in the data that I was trained on.”