Escaping Sameness and Creative Surrender in the Age of AI

by Robby Vaflor

Lately, I’m increasingly finding myself yearning for the old internet—think 2000s to early 2010s—when you actually surfed the web. It was a time when user activity felt more open-ended, and people weren’t just funneled into the same handful of platforms. While many of the websites then (even branded ones) lacked polish, they certainly weren’t lacking in character: fun little mini games, easter eggs, animations, and chunks of text you actually wanted to read.

The last example in particular, is something that I have been thinking a lot about. Having written online content in some form for over a decade, I’ve seen how my professional approach to writing has shifted from youthful hubris to complete surrender to the search engine optimization (SEO) playbook. As I write this article, I’m fighting the urge to optimize the content for SEO—what was something I had to learn when I transitioned from tech journalism to the corporate domain of marketing is now muscle memory, and it takes a conscious effort to unlearn.

Back when I started writing in 2015, the transformation of the digital space was well underway. Our attention was being consolidated, digital ads dominated website layouts, and the pecking order that is search rankings had been established—the wild west of the world wide web was over. Creating content for websites became primarily focused on what search engines think people would find value in based on historical data of search queries. Writers then became digital marketers, prioritizing keyword density rather than original or unique writing. Styles adapted to become something that can easily be read by search engines rather than people.

Suddenly, the internet was filled with SEO slop: not so much articles, but rather hollow shells in which to stuff keywords. Corporate websites were not the only ones guilty of this—those fun, quirky little personal pages that I alluded to had to play the game in order to have a chance at ranking. Even our newsroom had to learn to make our journalistic writing search engine-friendly. As time went on, SEO “best practices” became non-negotiable. While search engines such as Google have noted that original, helpful content is what makes websites rank, with how content creation becomes an industry where pages related to a keyword turn from thousands to millions, quantity ends up beating quality. It’s the same thing for multimedia content: marketing teams’ quest for virality meant images and video were also subjected to the same formulaic approach—”what worked for them might be what would work for us.” And what once was a bright, colorful place filled with endless possibility, the web has now been doomed to the fate of sameness.

AI is Speeding Up Sameness

Unfortunately, it doesn’t seem like the sameness problem will be going away any time soon. In fact, as AI becomes more integrated into search engines, social media platforms, and in content creation, there is growing concern that this technology is speeding up sameness.

In a way, an AI model is an aggregate of all (available) data on the internet. Through billions of samples, it is able to develop an amalgamated understanding of nearly every topic that exists. However, this process that gives it so much potential is also its key limitation.

Generative AI creates “new” content based on consensus—what’s the most common answer is most likely the right one. In this sense, AI is not so much a repository of hard facts, but rather an echo chamber of the most dominant perspectives. So when you ask ChatGPT to write you a blog post, it will give you an article that was not necessarily written in the best way possible, but the most “average” way possible.

Think about all those keyword-stuffed articles about nothing that I mentioned in my confessed complicitness in search engine-first writing. They are floating around in AI datasets with countless other similar articles written by thousands of writers, to be fed to AI models along with every “SEO best practices” article on the internet. How big of an influence would that have in AI’s consensus-based decision-making? If you ask ChatGPT about writing professionally, what would be its parameters on choosing the advice to give? Unless your prompting is accurate and precise, you’re going to get the average consensus; certainly not groundbreaking ideas. And once this AI-generated content based on faulty reasoning goes out into the world, it will be fed back into the same AI model, reinforcing false assumptions and bad practice.

Surrendering Judgement to AI

This is not to say that AI isn’t useful. In my personal experience of using it at work, it has been very helpful as a jumping off point for research and ideation. I’ve also used it to conceptualize designs, checking to see if the concept works, and then creating a completely original design on my own. What is concerning is the speed at which it has overtaken human thinking in many scenarios, especially in creative work. Browsing social media and company websites, I see many instances where humans have completely surrendered all judgement and creative decision-making to AI—some of the most egregious examples include AI slop and brain rot, and copy-pasted, AI-generated text with placeholders (e.g. [Your Company Name], [Insert Example]) still intact.

As mentioned earlier, AI is not immune to error or mislearning. Just as us humans can fall into a misguided groove (like prioritizing SEO over audiences), AI can be taught to follow flawed reasoning if it was fed enough bad information. This means that to responsibly and effectively utilize AI, we must scrutinize its output with more or the same rigor that we would for human output.

Last year, Ignited Word’s founder and principal, Malaika Cheney-Coker, wrote an article about safeguarding human cognition and creativity in the age of AI, noting that it should be a systemic effort to ensure cognitive potency of the human workforce, making sure that we continue to stoke human cognition and creativity as we adopt AI. Examples of how both individuals and organizations can protect against creativity erosion are numerous. For example, individuals can use AI as, “a mental sparring partner.” When about to query for some creative ideas, use your own “slow think” first and see if you come up with better ideas than the bot. “If going straight to the bot for ideas, challenge yourself to come up with something missing from the list – or a unique iteration of each shortlisted idea.” 

To add to this, organizations should invest time and resources to develop AI literacy among their workforce; many were quick to jump the gun on AI adoption without properly equipping team members with the necessary knowledge to better understand AI’s capabilities, limitations, and tendency to erode human creative capacity.

Rebelling Against Sameness

With the integration of AI into search engines, it has been observed that visibility for the majority of websites on the internet has further decreased. Coupled with the introduction of sponsored results, and the pre-existing bias of search engines to show results from only the biggest websites (Facebook, Reddit, Wikipedia, etc.), there is very little room for much smaller websites to be visible. Paid advertising is ineffective up to a certain point of spending, and organic marketing strategies are not the same as they were five or ten years ago.

The prevailing wisdom would be to stay the course, but what does that look like in 2026? Should we continue to chase metrics rather than forging deeper connections with our audience? If the traditional SEO approach is no longer offering results, then perhaps it’s time to bring back original thinking: to experiment with different channels and storytelling approaches. Maybe it’s time to put less emphasis on SEO and start writing and creating for people again. Maybe we should start thinking about introducing more life, creativity, and interactivity into websites—to bring back the quirky games and interesting easter eggs. Maybe we should go back to writing with rhythm and intent, rather than following SEO best practices. Maybe we should stop trying to copy trends and content formats, and start exploring our own. Wouldn’t it be better to create for and be remembered by human people instead of search engines?