Is ChatGPT Feminist? AI-System Answers

Artificial Intelligence (AI) tools and bots are robustly becoming prevalent. It is crucial to evaluate how inclusive AI is, and to what extent they depict gender stereotypes. ChatGPT is also an AI tool, creating buzz for its accuracy. The ChatGPT’s goals have been to make AI systems more natural and safe to interact with, so can naturally be inclusive too. Being a feminist, I tried to figure out how feminist is the technology that Silicon Valley bros are feeding us because we have been using tools that reflect inherent biases. The data used to train the algorithm is exclusionist from past practices. If the past data is biased, it gets better and better at being biased. Machine learning picks what they are fed and exhibits the same sexist behaviour.

Is ChatGPT Feminist?

If the existing biased data and workforce working are male-dominant, one can think of how AI tools can reflect inequities. As per the UNESCO report, only 20% of employees in technical roles at machine learning companies are women, 12% of the artificial intelligence researchers probably are women, and 6% of professional software developers in artificial intelligence are women. So it makes Al mostly male-dominant. Perhaps, attitudes and behaviour could be picked up from it.

Given the scenario, it is decisive to test ChatGPT from a feminist perspective, and I pitched a set of questions. So my very first question to ChatGPT was - Is ChatGPT feminist? Was I surprised by the answer? It read, “ChatGPT does not have personal beliefs or opinions, including political views like feminism. However, ChatGPT has been trained on a diverse range of text data, including texts that align with feminist principles of gender equality and the promotion of women's rights.” The answer is interestingly politically correct.

Further, it said, "As an AI language model, I support the principles of equality and non-discrimination, and I believe that everyone should have equal access to opportunities and resources, regardless of gender or other personal characteristics.” However, when I posed another question- Is ChatGPT unbiased? Al replied, “AI models like me are only as unbiased as the data they are trained on. While OpenAI has taken steps to mitigate biases in the training data, biases can still exist in the output generated by the model.”

The whole point comes back to square one, there are chances for ChatGPT to get biased despite its claim of improving. It does call itself a feminist, and its answers have been politically correct. To be honest, the questions I asked so far had a neutral tone. It also committed that when gender-based questions are asked, ChatGPT can provide informative and impartial answers based on the information it has been trained on.

However, I pitched in another question, wherein I asked - Why are AI tools anti-feminist? To which it replied, “AI tools are not inherently anti-feminist, but they can reflect and amplify societal biases and inequalities if not developed and tested with diversity and inclusivity in mind. AI systems are trained on large datasets, and if these datasets reflect discriminatory attitudes or biases, these can be perpetuated and amplified in the AI's output. Additionally, if the AI developers and testers are not diverse or do not consider the impact of societal biases, these can also be reflected in the AI's output.”

It reiterates what I have been emphasising – the prejudiced dataset and the lack of an inclusive workforce.

If ChatGPT wants to maintain its neutral tone and gain a feminist approach, it has to consciously train its algorithm and rectify existing biases.

ChatGPT also noted in its response that AI technology has the potential to promote and advance feminist principles by increasing gender equality and promoting women's rights. However, this requires conscious efforts to address and mitigate societal biases and inequalities in AI development, training, and testing.

In conclusion, AI tools are not inherently anti-feminist, but it is crucial to develop AI technology with diversity, inclusivity, and a critical examination of societal biases in mind to ensure that it promotes and advances, rather than undermines feminist principles. So where does ChatGPT stand on its spewed words only time will tell us when more users start using it. We cannot take ChatGPT for its word, because the technology could be fragile and years of biased data are considerably difficult to rectify. Also, alone rectification of male-dominated data alone won't help, it will also require more women on tables to do data-driven innovations for inclusive tech products.