Experimenting with ChatGPT

by diana

I was really intrigued by the parallels that Bethany Nowviskie drew between land stewardship and digital stewardship. As I learn more from Indigenous scholars like Kimmerer about our responsibilities to one another, I find myself reckoning with the interconnectedness of everything. I took this question from Nowviskie’s piece and decided to pose it to ChatGPT.

Prompt: Please respond to this question: What kinds of indigenous knowledge do we neglect to represent—or fail to understand—in our digital libraries?

Response:

As an AI language model, I do not have personal opinions or feelings. However, I can provide information that might be helpful to answer your question.

Indigenous knowledge systems include diverse and complex ways of understanding the world and managing natural resources, passed down through generations of indigenous communities. These knowledge systems are often closely tied to cultural and spiritual practices, and they have been developed and refined over thousands of years to ensure the survival and well-being of indigenous peoples.

Digital libraries have the potential to provide access to a vast array of information, including indigenous knowledge, to a wider audience. However, there are several challenges in representing and understanding indigenous knowledge in digital libraries. Some of the neglected areas include:

  1. Oral traditions: Many indigenous cultures have rich oral traditions that convey important knowledge and history. However, these traditions may not be represented in digital libraries, as they are not written down.
  2. Cultural context: Indigenous knowledge is often embedded within cultural practices and traditions that may be difficult for outsiders to understand. Digital libraries may not provide enough context to understand the meaning and significance of this knowledge.
  3. Intellectual property rights: Indigenous knowledge is often held collectively by a community, rather than by individuals. Digital libraries may not respect indigenous peoples’ intellectual property rights, leading to exploitation and appropriation of their knowledge.
  4. Language barriers: Many indigenous languages are endangered, and there is a risk that indigenous knowledge may be lost if it is not documented in these languages. Digital libraries may not have enough resources to support the preservation and representation of indigenous languages.
  5. Western bias: Digital libraries are often developed from a Western perspective, and may not incorporate indigenous ways of organizing and representing knowledge.

Overall, to ensure that indigenous knowledge is represented and understood in digital libraries, it is crucial to involve indigenous communities in the development and curation of these resources. This can help to ensure that indigenous knowledge is represented accurately and with appropriate cultural context, and that the intellectual property rights of indigenous peoples are respected.

Reflections: I was struck first by how quickly so much information was generated. I also kept expecting the response to end…but it kept going! This response feels so thoughtful, which got me thinking about the care that Nowviskie included in her original question to begin with. I see so much potential in the possibilities of learning with AI, and believe that interactions with it are a sandbox for us to learn and metabolize a great deal of abstract or concrete information quickly, and also an opportunity for the AI itself to learn about the questions that are important to us. I am curious about the ethics behind using AI tools; this information came from somewhere. Who gets credited? Are certain kinds of knowledges privileged? How are they protected? As a middle school educator, I wonder about how we might teach students to use tools like this responsibly.

1 thought on “Experimenting with ChatGPT

  1. Katina Rogers (she/her)

    This is really interesting, d. I’m struck by the fact that you didn’t ask the bot a question about feelings, and yet it interpreted it as such. Why does a question about Indigenous knowledges get misconstrued as affective rather than intellectual, and what does that imply about the way the language model interprets Indigeneity? And yet, as you mention, it goes on to offer a pretty solid response, more nuanced than I would expect.

Comments are closed.