Dear Dub: Is AI a Feminist?
To a girl who uses technology in the most average way: to watch TV, see in the dark, and use the GET app, the concept of AI was incredibly foreign to me. In fact, it had not really been on my radar until the end of last year. Before that, the only exposure I thought I had to AI was Siri, who we all recognize to be famously bad at the job. And even then, Siri was more of just a silly little feature on my phone and not something I could potentially rely on in any legitimate capacity. However, these days, Siri cannot compete with the free and widely available source of artificial intelligence on the internet. There are hundreds of websites that can generate images, songs, essays, answer questions, tell (actually funny) jokes, etc., with the click of a button.
This, of course, is objectively really cool. The idea that I could hunker down and conjure computer-generated image after image while procrastinating my homework was something I held dear. Because of how seemingly simple these intelligences are at doing the things we ask of them; one doesn’t organically question how something like this is possible. However, I had a sneaking suspicion that I wouldn’t like what I found out, and I was right. This is because every time you use AI, you receive a response amassed from hundreds and thousands of pieces of information and art that is on the internet. So, in other words, AI is actively plagiarizing any/everything that has ever been electronically published, which is so super not cool. Don’t believe me? Let’s ask the beast itself:
Grace: Where does the information you know come from?
ChatGPT: The information I know comes from the extensive dataset that was used to train me. This dataset comprises a wide variety of text from the internet, including:
Books: Excerpts from a wide range of books on various subjects.
Websites: Text from websites, articles and online sources covering different domains.
Research Papers: Scientific and academic papers.
News Articles: News reports and journalistic content.
Encyclopedias: Entries from encyclopedic sources.
Forums: Text from online forums and discussion boards.
In other words, quite literally anything it can get its figurative hands on. You may think that it is innocent to ask ChatGPT a question and rely on it as a resource for producing writing. There is this façade that AI presents where whatever you want to produce, say a poem or short story, is generated in seconds, quite literally conjured out of thin air and entirely random. However, the formulaic style of writing in which the medium is written, is entirely concocted by modeling thousands of real people’s work, who don’t get credit for it. You wouldn’t think it morally right to take all of your classmate’s essays and use them to make one super essay, so why is it okay when AI does it?
In writing this Dear Dub, the way that I view technology and artificial intelligence has really changed. Before I did not see it as a threat to individual rights, just a tool to better allow universal access to knowledge. However, I see that at the rate artificial intelligence is evolving, it is directly affecting people’s individual freedoms and claims to their own production. We have been using AI for years without it being something we talked about or wondered about the morality of. It is only now that it has gotten to the point of no return that questions have begun to be raised. The protection of people’s work, their creativity and their brilliance, should not be so easily discarded and minimized by our society. It is not an innocent thing for the people behind AI to exploit (and in some cases capitalize on) the individually published work of real people. Given that this is inherently an issue of the human right to autonomy of their own mind and creativity, I personally consider this to be a feminist issue. In my mind, to be a feminist means actively supporting and protecting individual personhood and identity. To actively disregard this is to violate that right – leaving me to believe that AI is not a feminist.