Cultivating an Aligned AI
- olivia1445
- Aug 20
- 1 min read
Lately I’ve been diving into my AI chatbot — not just using it, but really looking at the biases baked into how it was designed, trained, and what kinds of data it privileges. What I found is worrying: it amplifies many of the well-known biases, but also some of the more insidious ones — patriarchal systems, western dominance, under-representation of women’s issues, erasure of Indigenous knowledge, materialism over spirit, productivity over humanity.
At first, dipping my toe into this bottomless ocean left me feeling pretty overwhelmed.
But what pushed me into action were two things:
AI isn’t going anywhere. By 2027 it will likely be embedded in almost everything. It’s neither good nor bad in itself — it’s about how we choose to use it.
My kids. I watched them use ChatGPT like it was an all-knowing Oracle — trusting implicitly, without question. They didn’t know its limitations until I explained them, and for schoolwork the path of least resistance quickly becomes: “Ask ChatGPT.”
So, I pulled together a simple guidebook to help people understand what these biases are, why they matter, and what we can actually do about it. Inside are copy-and-paste prompts to help you tweak your own chatbot, plus some reflections on asking better questions and engaging with AI more consciously.
If you’re curious, I’m happy to talk it through — and please feel free to share it on. The more of us using these tools with awareness, the better.
Comments