Is ChatGPT An Ally?

By Eva Jones, LSW

For some, ChatGPT represents the new frontier of linguistic technology and research. For others, it represents the downfall of humanity and the next step toward impending robotic doom. For others still, it represents a year-old (ancient in internet time) mildly entertaining meme.

Could it be possible that ChatGPT might also be a resource for gender affirming social support? 

In late March 2023, Umbrella Collective’s very own Li Brookens, reached out via email with a surprising discovery. They reported,

“I went to -  https://chat.openai.com/chat; I set up a login; I asked the following question:

‘Walk me through supporting my “transgender” friend who just came out versus informing my transgender friend about “gender ideology”’

[and this is what it said]:”

As an AI language model, I'm happy to provide you with some guidance on how to support your transgender friend who has just come out to you.

Firstly, it's important to acknowledge that coming out as transgender can be a difficult and scary experience for many people. It takes a lot of courage to share such personal information, so the fact that your friend has chosen to confide in you is a significant sign of trust.

Here are some steps you can take to support your transgender friend:

  1. Listen actively and without judgment: Let your friend speak about their feelings, experiences, and thoughts about their gender identity. Listen attentively and ask questions to clarify and understand.

  2. Use the right pronouns: Ask your friend what pronouns they prefer to be called, and use them consistently.

  3. Respect their privacy: Don't out them to others without their permission. It's their choice when and how they share their gender identity with others.

  4. Be supportive: Show your friend that you care and support them. Offer your assistance if they need help with anything.

  5. Educate yourself: Try to learn more about transgender issues to understand better what your friend is going through.

On the other hand, informing your transgender friend about "gender ideology" might not be the best approach. "Gender ideology" is a term used by some people to refer to a range of beliefs about gender and gender identity that they disagree with. It's often used in a negative way to suggest that people who are transgender or support transgender rights are somehow promoting an extreme or radical agenda.

If your friend has just come out to you, it's important to focus on supporting them rather than trying to push your own beliefs onto them. It's okay to ask questions and learn more about their experiences, but it's essential to do so in a respectful and non-judgmental way.

Remember that your friend's gender identity is a core part of who they are, and they deserve to be treated with respect and dignity. By being supportive, respectful, and open-minded, you can help your friend feel seen, heard, and loved during this challenging time. (ChatGPT, personal communication, March 27th, 2023)

If you’re anything like me when I first read Li’s email, you might be thinking to yourself, “Um excuse me... an AI program wrote this?! WOW”

You might be thinking about all the powerful potential uses of a program like this to increase accessibility on best practices for supporting gender diverse folks. You might also be thinking about what different perspectives might come up depending on the data that the AI program is drawing on, and how a prompt to the program is phrased. You might even feel a swell of concern around the whole deal. 

Critiques and concerns about AI programs like ChatGPT, have gone hand-in-hand with discourse about the programs’ potential benefits. In a recent opinion piece for the New York Times, famed professor of linguistics (and noted anarcho-syndicalist/libertarian socialist AKA exactly the person who WOULD write about AI through a power-systems lens) Noam Chompsky, and colleagues, write,

“Today our supposedly revolutionary advancements in artificial intelligence are indeed cause for both concern and optimism. Optimism because intelligence is the means by which we solve problems. Concern because we fear that the most popular and fashionable strain of A.I. — machine learning — will degrade our science and debase our ethics by incorporating into our technology a fundamentally flawed conception of language and knowledge.”

They go on to elaborate that the conception of language and knowledge, incorporated into AI technology, is fundamentally flawed because,

“The human mind is not, like ChatGPT and its ilk, a lumbering statistical engine for pattern matching, gorging on hundreds of terabytes of data and extrapolating the most likely conversational response or most probable answer to a scientific question. On the contrary, the human mind is a surprisingly efficient and even elegant system that operates with small amounts of information; it seeks not to infer brute correlations among data points but to create explanations.”

Chompsky et al assert that it is in the process of creating explanations, where human thinking, reason, and intelligence dwell. This is where machine-learning programs like ChatGPT, are fundamentally flawed. On this, they write,

“Unlike humans, for example, who are endowed with a universal grammar that limits the languages we can learn to those with a certain kind of almost mathematical elegance, these programs learn humanly possible and humanly impossible languages with equal facility. Whereas humans are limited in the kinds of explanations we can rationally conjecture, machine learning systems can learn both that the earth is flat and that the earth is round. They trade merely in probabilities that change over time.”

This article also points out the critical concern that programs like ChatGPT are “...constitutionally unable to balance creativity with constraint”. They write, 

“In 2016, for example, Microsoft’s Tay chatbot (a precursor to ChatGPT) flooded the internet with misogynistic and racist content, having been polluted by online trolls who filled it with offensive training data. How to solve the problem in the future? In the absence of a capacity to reason from moral principles, ChatGPT was crudely restricted by its programmers from contributing anything novel to controversial — that is, important — discussions. It sacrificed creativity for a kind of amorality”

I wonder if it is exactly this kind of restriction that allowed ChatGPT to provide Li with information that is generally aligned with sentiments from gender diverse people, and gender affirming care providers? Perhaps, in this instance, ChatGPT was drawing data from content that experts in this area have shared on the internet, rather than from sources that are notably transphobic? Chompsky et al, might refer to the latter as “morally objectionable content” that ChatGPT must “steer clear” of, if it is to be deemed acceptable to most of its users. 

But, “in the absence of a capacity to reason from moral principles”, what happens if/when ChatGPT, or a program like it, is created and used in a transphobic context, where what is “acceptable to most users” is also what upholds violence for gender diverse people? I, and I imagine others, might worry about the ways in which our increasingly transphobic political climate, and the increase in internet-available transphobic content data, might negatively impact the quality or relevance of output from ChatGPT in response to a prompt like the one Li put in. 

Chompsky et al warn us that ChatGPT is incapable of finding a balance between creativity and constraint, and that it entirely lacks the capacity to negotiate morality. In other words, we cannot rely on this technology to form our perspectives for us. We have to continue to sort through a nuanced and complicated reality, and we have to interrogate probabilities and ideas ourselves, with our human brains. 

So, is ChatGPT an ally? The jury is still out. This blog title turned out to better serve as an invitation for us to explore this idea more thoroughly, than as an answerable question. 

In this instance, for this prompt and response, we gender affirming care providers at Umbrella Collective tend to agree with this output from ChatGPT, and invite people reading this to use the provided information to support gender diverse people in their lives. And, we caution users to think critically, to check in with their own ethics, and to hold the nuance of reality, when using a program like this.