Beyond the algorithm

By Herman M. Lagon 

Let us face it. As students, professionals, or creatives, most of us have already been exposed to digital tools for writing. Whether it be grammar checkers or content generators — the support is a click away. This is not anything new — each decade has had its tech breakthroughs, from scientific calculators in the 80s to computers in the 90s, and then, there was the internet surge in the early 2000s. In the 2020s, we now have AI-powered writing tools (e.g. ChatGPT, Grammarly, OpenAI, Perplexity, Jasper and Gemini / Bard and Jenni and Rytr)–the likes of which nobody has ever experienced. Yet with these tools comes a subtle danger: losing that unique, human voice in our work.

There are a few red flags to steer clear of for anyone trying to avoid sounding like a robot. Some common phrases, like “delve,” “facilitate,” “tapestry,” or “maximize,” seem innocent enough but can often hint, albeit not perfectly, at AI authorship. ChatGPT and similar tools lean on these words because they are safe, broadly applicable, and easy to work into almost any sentence. Unfortunately, these phrases feel repetitive over time, giving AI-generated text a specific hollow tone. The more we avoid such catch-all words, the better our writing sounds—fresh, exact, and ours.

Another telltale sign of AI-generated text is the predictable structure it follows. It is often overly formal, with phrases like “additionally,” “consequently,” “moreover,” or “ultimately” peppered throughout. Nothing is wrong with these words, but when overused, they start to sound stiff, as if the writing follows a formula. Real human writing rarely sticks so closely to a script, often flowing with a mix of short, punchy phrases and longer, thoughtful reflections. Mixing sentence lengths and inserting conversational transitions can make writing feel personal and relatable—qualities AI still struggles to capture.

The reliance on overused adjectives like “dynamic,” “robust,” “profound,” or “efficient” also makes AI-generated text stick out. Consider the word “innovative.” When everything is described as “innovative,” the term loses its spark. Instead, we should choose descriptions that zero in on specifics, like “fresh,” “thoughtful,” or “groundbreaking,” each carrying its nuance. It is about choosing words that mean something specific to our point, avoiding generic terms that feel like filler.

Here is another way to personalize our writing: targeted anecdotes and analogies. While AI can generate a range of examples, it is less adept at weaving in real-life moments or relatable metaphors. For instance, if I were talking about how important it is to sound genuine, I might mention that AI writing sometimes feels like the overly scripted conversations we hear on sitcoms—not quite lifelike, just a little too perfect. These personal touches add depth, making our work sound authentic rather than algorithmically optimized.

Using local expressions can be a great way to sidestep the AI trap for us, who often speak various languages and dialects. Whether it is a Pinoy idiom, a Hiligaynon phrase, a “po” to end a sentence, or simply the rhythm of how we naturally speak, adding a bit of home-grown flair makes our writing uniquely ours. Instead of saying, “It’s important to note,” we could use something like, “Remember, folks,” bringing warmth and relatability that is harder for AI to imitate. Local color is not just about language—it is about context, too, like referencing our passion for karaoke to emphasize a point about overcoming stage fright. These are details that keep our voice at the heart of our work.

Let us be honest and blunt: AI helps organize ideas or get a first draft done quickly. But if we use it for everything, it is like expecting a calculator to write an essay. As we know from school, while the calculator is great for speeding up arithmetic, you must learn to think mathematically. In the same way, we need more than AI to teach us the artistry of crafting words that touch readers. So, we can use AI, but it should only be the spark—not the whole flame—of our creativity.

There is also the trick of varied prompts when working with AI. Instead of asking for “an article on climate change,” try “explain why typhoons are getting worse in the Philippines, and include specific examples of recent storms.” By guiding AI with specific, personal angles, we are more likely to generate content that sounds rooted in experience and knowledge rather than generic concepts.

One practical tip to catch any lingering “AI-speak” is to read aloud. If a sentence sounds too stiff or overly polished, it needs rewriting. AI tends to overuse specific patterns, so hearing the words can highlight where we need to simplify or add a personal touch. Writers have used this method for generations, long before digital tools were around—our ears are great at picking out awkward or “libagon” (dissonant) phrases.

The last trick to beating AI-sounding text? Have fun with it. We Pinoys love to play with words and add a bit of wit and humor to our conversations. Bringing that same playfulness into writing makes it feel alive, like there is a person on the other end of the page. When we treat writing as a living conversation instead of a mechanical process, we do not just avoid sounding like a bot—we add that spark of originality that only comes from genuine expression.

So, as AI evolves, the question is not whether we will be able to keep using these tools (we will, inevitably) but how we can use them wisely. Like calculators, computers, or the internet, AI has its place as long as we remain in control. But the true magic of writing is that it is personal, crafted, and endlessly surprising. Let us keep it that way.

***

Doc H fondly describes himself as a “student of and for life” who, like many others, aspires to a life-giving and why-driven world grounded in social justice and the pursuit of happiness. His views do not necessarily reflect those of the institutions he is employed or connected with.