The Last Blog

A final reflection on how my blog evolved over the semester, what changed in my writing, which posts challenged me most, and how my perspective on responsibility, ethics, and technology shifted through revisiting them.

Most Informative: “Harms in Machine Learning”

This blog is definitely the one I learned the most from. It was the first time I realized that bias in AI isn’t one failure, it’s a chain of tiny decisions that stack up into something with real world effects. Writing it pushed me beyond: “this model is biased” into: “here are seven different reasons it becomes biased, and here’s how I’ve contributed to each of them in my own work.” I remember slowing down while writing this post and thinking, “Wait, we literally did every one of these things wrong in that one project we thought was simple.” That realization was uncomfortable…but important.

When I first wrote this post, it was almost entirely descriptive, more like a summary of what the article said. After revisiting it, I realized that I wasn’t actually connecting those concepts to my own understanding or experience. In revision, I added examples from my past experience where I had unknowingly contributed to the exact biases described, and comparisons that made the concepts feel real. I also clarified the structure so each form of bias connected to a specific story or implication. That revision shifted the post from “explaining the article” to actually thinking through its consequences, which made it feel more honest and meaningful.

Most Surprising: “Complete Delete”

I didn’t expect this blog to shake up something so ordinary. The word “delete” used to feel final. Until I learned it wasn’t. Writing this post made me think differently about something I do every day, especially when it comes to privacy. I stopped assuming systems work in my favor and started noticing how they are designed to prioritize storage, convenience, or retention rather than user clarity. There was one moment that stuck with me: pairing your phone to a rental car and then just driving away, leaving your call logs, messages, and sometimes navigation history saved for the next person. That’s the moment when “delete” stopped being theoretical and became personal.

Understanding My Writing Process

I didn’t realize it until recently, but this blog format made me write differently than I do in academic work. There is something freeing about writing for an online space, where transitions don’t have to be rigid and tone isn’t forced to sound formal. What I think changed the most:

• I became clearer without needing to sound complicated.
• I started using examples from my own life.
• I stopped summarizing and began translating ideas into things that mattered to me. One of the hardest parts was structuring each post uniquely. Essays come with built-in rules. Blogs don’t. And having to design my own structure each time was frustrating, but also the thing that made me better. Honestly, I prefer this format. It feels like thinking out loud, but with purpose. As for patterns, I noticed that: I always start with something relatable and I end with unresolved questions instead of neat conclusions. I like that about my writing now. Not everything became easier, though. Sometimes I caught myself trying to sound polished instead of honest. Letting go of that was its own form of growth.

  Connecting to Our Course

  My favorite topics to write about ended up being the ones that touched real people—moderation, algorithmic persuasion, hidden labor behind AI systems, and the emotional implications of machine learning. The posts that stayed with me weren’t the ones about how models work, but rather what happens when they do work. For example:

• When personalization becomes manipulation
• When deletion becomes impossible
• When automation masks invisible human effort
• When AI reshapes what fairness even means

There was also a specific moment when course ideas “clicked”: Writing about deployment bias made me understand that a model being “correct” isn’t enough, what matters is how people interpret and use it. That realization turned ethics into something ongoing, not something that happens at the end.

Looking Forward

Now that I have a website I actually enjoy looking at, I want to keep it alive, not necessarily as weekly posts, but as a place to organize things I learn. I want it to become a reflection space, maybe even a place to write about tech in a way I don’t get to in class. As for writing identity, I don’t think I want to be purely academic or purely technical. I like the middle place, the space where you can say, “Here’s what this tool does, and here’s how it feels to use it.” That’s the tone I want to develop more. If I set one writing goal for next semester, it would be consistency. Not long pieces, just small reflections, even if they’re messy. Writing helps me make sense of things in real time, not after the moment passes. And if these posts reminded me of anything, it’s that writing is not just documenting learning,it actually creates it.

Final Reflection

Looking back, I’m surprised by how much these topics have changed the way I think about technology. Not in a dramatic “I will never write code the same way again” way, but in a quieter shift: I notice things I used to ignore: who gets left out of datasets, when systems encourage convenience over autonomy,when metrics flatten real human behavior, and most importantly I started questioning For Whom?

And I notice when something that feels harmless, like deleting a file, is actually layered with assumptions, limitations, and consequences. I think that’s the real accomplishment, not writing polished posts, but walking away more aware, more curious, and more responsible. And that feels like the right way to end this.