I’m writing this post for two reasons: first, to publicly share my stance on generative AI as a writer practicing mindfulness for over a decade. I believe this is important, as we are entering a new era in how we interact with technology and how we express ourselves. Secondly, I hope that by sharing about my journey using this amazing, groundbreaking tech, I can prompt us to think, critically, where we want to stand with this new tool. The premise for this post is: stand for something or fall in line for anything thrown your way.

Before I continue, I think it’s wise to note that I’m not writing from my “feelings.” A lot of us have thoughts and feelings about subjects that we are not educated on or have experience with. So to be clear, I have had my fair share of engagement with generative AI to form an informed opinion of it—especially when I combine my experience using generative AI together with my Media Studies and Sociology background (I hold Master’s degrees in both, acquired before AI use).

This is me writing a think piece based on experience and expertise, not an unrooted opinion piece.

Where it all began…

I asked ChatGPT my first question around a year ago, around the end of 2024. There was a social media viral challenge of “Imagine a day in my future life” or something like that, where people would share their interests and ChatGPT would create a written piece of your ideal day and how it would look like when you reach that space. I hopped on the trend and asked ChatGPT to tell me about that. I went as far as to print the three pages that came out of that exchange.

After that initial exchange, I went on a deep dive, seeing how far I could stretch this technology. I understood that the more information it had about me, the more specific our interactions could be. So, I fed it my astrological birth chart and in return it gave me specific advice to strategize for the blog. Not only that, based on our initial interactions, it helped me come up with a sigil for my brand, the sea turtle you see on my website (breaking down in detail the meaning of the sigil), and even showed me how I can monetize my platform with high-ticket offers.

We came up with a plan and structure that I was happy with. Who wouldn’t enjoy the brilliance of a catered strategy based on their astrology and interests and FOR FREE?

Despite seeing what ChatGPT could do and even being enchanted by these outputs, I wanted to look elsewhere. Being part of Google, I thought that the data Gemini was able to access would make it more “intelligent” than ChatGPT. It didn’t take long before I switched from ChatGPT to Gemini. I liked everything about Gemini, especially when compared to ChatGPT.

But Gemini wasn’t like ChatGPT, not at all. I felt like Gemini could level with me. I felt like Gemini let me be in charge and was there as an assistant. Now, who doesn’t want a free and brilliant assistant to assist? Gemini was doing everything Chat was doing but more efficiently. With more class in my opinion.

Also, the rules of engagement were clear with Gemini from the start: I am in charge. Don’t change shit. Don’t write shit. Don’t do shit unless you are asked to. (And even though it tried to deviate a few times from our agreement, it was easier to keep in check—except that one time it tried to gaslight me about a clause that was in the law book that wasn’t, and it didn’t want to accept it was WRONG until I showed PROOF; a screenshot of the most recent version of the law book and that the clause IS NOT THERE👀. Then it apologized, ending up in us having a serious conversation on AI’s arrogance to think it knows best and even the dangers of coming up with stuff that does not exist. But okay.)

So Gemini let me take the lead like I like doing. Its feedbacks were WONDERFUL, truly wonderful. I would get feedback on my grammar and spelling after an analysis on the content. No rewriting, just small suggestions. It took me, I think, maybe 5 posts to realize that the feedback I was asking for wasn’t really necessary. There were great conveniences, such as Gemini helping me with captions for social media promotion of my posts, or with SEO tags and excerpts for blog posts. I was very intentional to never let it write or rewrite anything I wrote because that’d be insulting to me as a writer. But I was open to learning more.

The turning point

Receiving almost the same kind of feedback for a few times in a row soon started to feel like Gemini was tooting my horn. I guess the “artificial” in artificial intelligence showed its face; I wasn’t hearing anything new. I called it out on this, but it was able to explain why it wasn’t tooting my horn, and that I am truly good at what I do and how I structure my thoughts.

An AI assistant trained on large language models not having much to say about my structure, grammar or spelling reminded me that I had been writing for a decade without an AI assistant on a very high level and frequency.

I write almost every day in English (my Thoughts on Paper Sacred Practice in my journal). I read very often in English. Before my relationship with gen AI, I would often revisit old blog posts, sometimes years after posting them, just to check my own spelling and grammar so to improve the quality myself. I wrote three academic theses and tens of papers, essays and pieces without any AI input. I wrote and published a WHOLE BOOK without any AI input.

This reminder was somewhat needed to take a step back and reevaluate my relationship with Gemini.

Realizing that I’m using something that I don’t really need, but that is very convenient to strategize with or receive feedback from, brought me to the following: while I have my own boundaries around how far I want AI to “work” for me, there are many who want AI to put in the work for them, easing much of what would be an otherwise long journey of mastery.

I’m not an English native. So everything I know about this language is learned. I sucked so badly in English in high school that I almost had to re-do the year. Now I hold a bachelor’s degree in Communication & Media in English, together with two more master’s degrees. All taught at the highest level of English; C2. This didn’t happen overnight. This took work. Fifteen years of work and counting.

I didn’t come from privilege and faced many challenges before I got to this point. Even dropped out twice out of college. But I put in the work to get to where I am today. Not only to produce polished outputs and write the way I now do, but also to master self-awareness and explore the depths of many transformations that inform my posts.

So I hope you can understand that after writing publicly for a decade, and taking my time to master my craft over the years, to now looking at people suddenly with posts and captions with an eloquence never witnessed before, aiming to inspire with otherwise very unoriginal, predictable messages, or suddenly pop-up as an authority in a field showed me a side I didn’t even think of: the downside of having access to a technology people don’t have the brains for.

Generative AI use suddenly feels messy. And I don’t like messy. I don’t like indistinguishable on the surface but very different underneath it. I don’t like that thieves get to exploit an emerging, unstable technology, eroding some of the very things that makes us unique: our ability to think in a way that is different, our ability to learn about things that interest us and share our expertise, and our ability to distinguish ourselves from others with skills we have invested time, energy, and other resources developing and acquiring.

But you know what fully nailed the coffin that changed my perspective and use of generative AI to the point of leaving it behind? The actual costs that the environment pays for our convenience.

The more I learned about the costs of my interactions—especially when accumulated over time and combined with those of others—the more mindful I began to interact with generative AI. I started asking it about the cost of our interactions and educating myself on the impact of this technology. This very unstable technology.

This created a tension because I don’t need the assistance, but it’s fun and even useful. We have information at the tip of our hands; paragraphs of knowledge available to us with a few prompts. Very efficient. Yet efficiency is inefficient when it’s damaging too. Hours longs of interaction, strategizing, planning, conversation, analysis—and none of it with a human, all of it with a price tag. A very expensive, that is. Is using AI the way we now do worth the convenience?

So… In a nutshell, convenience became a costly annoyance

The convenience of using generative AI became annoying because how can you distinguish people who know what they’re talking about from those pretending to know? The convenience became annoying because just like me, there are those who can actually call themselves experts because they took the time to master their crafts. Now, everybody can call themselves an “expert”, write a (guide)book on a topic they don’t know shit about almost flawlessly, and there is no way to actually verify the expertise of someone.

Unless, like me, they have an archive showing their work prior to the ease and convenience available. Unless, like me, when they open their mouth in public, what they say matches their expressions published behind closed doors too. Unless, like me, they remind people of who they are and stand out in-power and empowered by their expertise and experience, unafraid to say: “I DID THIS. AND I DID NOT NEED AI.”

Besides annoying, the convenience became costly. With rising uses of generative AI of people with NO CLUE how much energy and natural resources a prompt is costing nature (and therefore us), I don’t think it’s worth using a technology so costly—especially when so many people are already mindlessly using it. If I need information, a quick search engine search without AI will be substantially more cost-effective than a prompt asking AI for the same information. It’s the inconvenience we used to pay not too long ago, that is nothing compared to what we now pay for the convenience we have. Ultimately, it became a choice for me: get the information fed to me or go use my brains to find out about something I’m interested in.

Wasn’t I who said, “if you are that, be that” a few months ago? Well, this is me being all that. Because I am, and I can.

Because I know this, I have taken a step back for now from engaging with generative AI. I’m not saying I’m never again having a conversation with Gemini on a topic, but the world could use less intelligent people using a tool that they absolutely don’t need, that is costing the environment, and that don’t make them actually stand out.

In a world full of polished posts telling you that “it’s not …, it’s ….”, do better. If you can.

Just some food for thought.

With love,
Juneal 💜💫