(ORIGINALLY WRITTEN BY NATHAN J. ROBINSON; Let ChatGPT turn you into a socialist)
Recently, I was attracted by new developments in artificial intelligence. The "text to image" generator can generate everything from stills from fake movies to imaginary new mash-up art genres to huge panoramas that imitate Hieronymus Bosch – AI is not only amazing in the scope of what it can create, but also in terms of "creativity".
But "generating fantastical images from text prompts" is just one of the extraordinary technological developments happening around us. AI's ability to use language has also improved so much recently that a Google engineer panicked that computer programs had developed individual wills (news last year).
Recently, there has been a lot of media coverage of ChatGPT, a language model made by OpenAI that is capable of natural language conversations on almost any topic. The New York Times said it was clearly the best AI chatbot ever released to the public, and countless images of people talking to ChatGPT went viral on social media.
The moment you start interacting with ChatGPT, you realize that Time is right: it's qualitatively different from previous chatbots. It doesn't try to mimic humans (imitation just appears to be a good data processor with no ideas of its own). But ChatGPT states clear, complex English sentences and understands even very complex issues well. If you let it explain what the core arguments of "against democracy" are, it will explain those arguments very well.
If you ask it to create a text-based game, or write you a paper on the history of bread, it will do so. It can write computer code, jokes, and dialogue from plays and TV shows. (I'm currently trying – with limited success – to get it to write a new season of Curb Your Enthusiasm so I don't have to wait for the show to return).
ChatGPT can provide decoration techniques as well as explain philosophical concepts. And it doesn't just paraphrase Wikipedia articles, it can do that if you ask it to write an article that compares and contrasts the views of two thinkers. It helps students with homework, which means we may be about to enter a world where homework cheating is so easy that homework assignment is meaningless.
Left-leaning ChatGPT
Since I'm a writer who has long been working on explaining political views, what excites me about ChatGPT is that it's very good at giving me the theoretical knowledge to help me persuade other people. For example, I wrote an entire book called Why You Should Be a Socialist, and ChatGPT provides more succinct, even persuasive arguments.
It can also be studied in more depth. For example, I specifically asked it to respond to the idea that under socialism, people would not be motivated to increase productivity. ChatGpt makes a number of arguments to counter this view, such as mentioning that the argument ignores that people are not motivated to work just for money. Other arguments and arguments will also be provided.
You can also ask ChatGPT to explain the "Socialist Critique of Liberalism" using the work of a socialist philosopher, such as G.A. Cohen (G.A. Cohen is a famous contemporary British philosopher and political theorist). I even asked ChatGPT to try writing a short utopian novel that gave me a story about life in a future socialist society. The things it creates, if not particularly memorable, are more satisfying:
(The translator has not translated the following screenshots, for viewing only)
Of course, I used the inducing question to ask it to advocate socialism, and you might wonder if it is equally willing to advocate for right-wingers. ChatGPT is programmed as an "anti-fascist" and cannot be induced to praise Hitler, but it can be induced to help promote theories. But when I asked it to write a surreal capitalist novel, it was clear that it was off topic.
It was asked to do a political orientation test and concluded that it was left-leaning. (It is said that its IQ test score is only 83 points [FYI]). In a way, though, it's a mirror that reflects the user's interests, for example, it doesn't tell a liberal user that he should read Marx and G.A. Cohen。
It can show you arguments for anything other than Nazism, which really makes it a "neutral" tool. (When I debated with ChatGPT whether it was really neutral or had implicit biases and positions, it got stuck in an endless loop). But for leftists who want to understand how another faction articulates their arguments and refine their own, ChatGPT is undoubtedly a great ally. I've read a lot of great articles written by ChatGPT, including: "Why Abortion Should Be Legal," "Why the Offspring of the Enslaved Should Be Compensated," "Why Ben Shapiro Runs Trains with His Mouth Full," "Why We Should Raise Taxes on the Rich," and "Why The Right's Stance on Climate Change Reflects the Conservative Politics' Threat to the Future of the Planet."
ChatGPT is not a good match for capitalism
Because I often write about these kinds of topics, you might think I'm worried that ChatGPT is about replacing me. But I'm actually happy. In fact, I don't like to explain these basic points repeatedly, and it would be better if a robot could answer people's questions without me having to do it. ChatGPT is not a demagogue of any ism, it is just good at helping people think about problems. Because I'm confident that the left's thinking is right, and I think the "good stuff" comes from a tool that helps people see and understand each side's argument.
There's one best reason to be a socialist that ChatGPT doesn't mention: because technology like ChatGPT threatens the jobs of many workers, only socialists can ensure that workers survive technological upheaval, especially when their labor market value is replaced by technology.
Otherwise, if the value of workers suddenly drops, just because of the endless stream of new robots, it will indeed be over. We find something like ChatGPT scary because it can already replace a lot of jobs and make life more difficult for a lot of people.
In fact, ChatGPT could have made our lives easier. Asked how a data engineer at Twitter could trick Elon Musk into thinking they were working hard, ChatGPT offered a pretty good set of advice. (Although when I asked what the magazine was supposed to write, it gave me some bland nonsense.) It seems that I cannot be completely "automated" yet).
And the fact that this doesn't mean that engineers can make their jobs easier [under capitalism] opens up a Pandora's box that makes them completely lose their jobs. And there are more dystopian scenarios, imagine an AI boss having complete control over your work.
It is important to ensure that new technologies bring prosperity to ordinary people
I'm skeptical of the theory of the upcoming "superintelligence," but I do believe that there are countless ways in which new technologies can make life miserable rather than better. These new AI languages make it clearer why we have to create a different society where a robot can't cause workers to lose their livelihoods just because they learn how to do something.
My adventures in AI art and text generation convinced me that even if humans aren't replaced, some very unfamiliar new technological capabilities are about to be unleashed, risking mass disruption.
AI hasn't started to impact music yet, but it's only a matter of time. But OpenAI (Manufacturing Image Generators DALL-E and ChatGPT) is already developing a "jukebox" that can create new songs in the style of existing artists.
For now, it's still pretty inferior, and the songs sound terrible (although its parody of Sinatra is coming up to standard, or if you're not a Bob Dylan fan, you might think the AI isn't bad), but I can foresee a future where anyone with a computer or phone can record a song that sounds decent.
We must ensure that new technologies bring prosperity to all, not exacerbate inequality. The new language model has some good applications, such as helping those who have difficulty in literacy to express themselves better and increase the popularization of knowledge and culture.
As someone with a lot of imagination but limited drawing skills, I'm most excited about being able to make cool images and being able to have a "universal expert" answer any questions that pop up in my head. AI can certainly help us achieve "fully automated luxury communism," although it can also go big wrong in countless ways.
We should look forward to the future, not fear it, because it should be a good thing to be able to create more and do more with magical new devices. As I said, the current ChatGPT is a tool that, if used effectively by socialists, can help advance the intellectual argument of socialism. But we are certainly on the verge of some kind of great change and should be prepared for social and economic change that we cannot even fully see.