Hey everyone,
Sam Ellis here, back on agntzen.com, and today I want to talk about something that’s been gnawing at the edges of my thought lately. Not the grand, sweeping philosophical debates, though those are always fun. No, I’m talking about the mundane, the everyday, the little choices we make that, when aggregated, define our digital lives. Specifically, I’m thinking about the quiet creep of predictive systems and the subtle erosion of our agency in the face of “convenience.”
My angle today isn’t about the robots taking over, or even the existential threat of superintelligence. It’s far more grounded: it’s about the increasing difficulty of simply *choosing* what you want to consume, create, or even think, when every platform, every app, every search query is already trying to guess it for you. It’s about the very real, very present challenge of maintaining our individual agency when the algorithms are designed to anticipate, influence, and ultimately, direct our next move.
The Echo Chamber of “You Might Also Like”
Remember when you used to browse a bookstore? You’d wander, perhaps pick up a book because the cover caught your eye, or a blurb hinted at something intriguing. You were the agent in that scenario, driven by curiosity, serendipity, or a specific quest. Now, think about your online experience. Whether it’s a streaming service, an e-commerce site, or even a news aggregator, you’re constantly bombarded with “recommendations.”
I’ve fallen into this trap more times than I care to admit. Just last week, I was trying to find a new independent film. I started on a popular streaming platform, thinking I’d explore their “indie” section. Within minutes, I was being shown films incredibly similar to things I’d watched before. Not bad films, mind you, but nothing truly *new*. It felt like the platform was saying, “We know you, Sam. You like these kinds of stories. Here are more of those.”
And for a moment, I almost gave in. It was easier. It required less effort to just pick one of the algorithm’s suggestions. But then I stopped. I thought, “Is this what I actually *want* to watch, or is this what the machine *thinks* I want to watch, based on past data?” The distinction, I realized, is crucial for our agency.
The Problem with Perfect Predictions
The goal of these predictive systems is to reduce friction. To make your experience “easier” and “more enjoyable” by giving you what you supposedly want before you even ask for it. On the surface, who could argue with that? Less time searching, more time doing. But what happens when the predictions become so good, so omnipresent, that they start to narrow your world instead of expanding it?
Consider this: if an algorithm always shows you content that aligns with your existing biases, you’re effectively living in an echo chamber. You’re not exposed to dissenting opinions, novel ideas, or even just different aesthetic sensibilities. Your intellectual and emotional space starts to flatten. The capacity for surprise, for discovery, for genuine intellectual growth, begins to shrink.
This isn’t just about entertainment. It’s about information, too. I’ve noticed it with news feeds. If I click on articles from a particular political leaning a few times, suddenly my feed is saturated with similar perspectives. It takes a conscious, deliberate effort to break out of that loop and seek out other viewpoints. And that effort is precisely what these systems are designed to minimize.
Reclaiming the “Unpredictable You”
So, how do we, as individuals, push back against this gentle but persistent nudging? How do we reclaim our agency in a world that increasingly wants to predict and pre-package our preferences?
1. Actively Seek Serendipity
This is probably the most important step. Don’t always take the algorithm’s first suggestion. Go off-piste. If you’re looking for music, try browsing by genre you rarely listen to. If you’re looking for a book, pick one from a different shelf in the physical library or use an online tool that specifically avoids personalized recommendations. The goal is to introduce randomness back into your consumption patterns.
One trick I’ve started using for news is to periodically visit the homepages of news outlets I don’t typically read, without logging in. This gives me a general overview of their main headlines, rather than a curated feed based on my past clicks. It’s a small act of rebellion, but it works.
2. Question the “Why”
When you see a recommendation, pause for a second and ask yourself: “Why is this being shown to me?” Is it because I genuinely expressed interest in something similar, or is it because the platform wants to keep me engaged, or because a particular product has a higher profit margin? Understanding the potential motivations behind the recommendation helps you evaluate its true value to *you*.
For example, if I’m on a shopping site and it recommends a specific brand of coffee, I might think: “Is this because I’ve bought coffee before, or is this a sponsored placement?” A quick search for “best independent coffee roasters [my city]” might yield a more authentic, less algorithm-driven result.
3. Cultivate Intentionality
Before you open an app or visit a website, have a clear purpose. Instead of just “browsing,” decide what you’re looking for. Are you looking for a specific piece of information? Are you trying to learn something new? Are you connecting with a particular person? When you approach your digital interactions with intent, you’re less likely to be swayed by the passive suggestions.
I’ve started a habit of writing down 1-2 things I want to accomplish online for the day. It could be as simple as “research topic X for 30 minutes” or “find a recipe for Y.” This small act of pre-commitment makes a huge difference in avoiding the rabbit hole of endless recommendations.
4. Use Tools for Deliberate Discovery
There are some wonderful resources out there that actively fight against the personalization bubble. Sites like AllTrails (not strictly AI-avoidant, but good for local, non-algorithm-driven discovery) or even just old-fashioned forums and niche blogs (like this one!) can be excellent for finding things based on human curation and shared interest, rather than predictive models.
For code snippets, consider how you might proactively seek out different perspectives in your development work. Instead of just hitting “enter” on the first Stack Overflow result, try to understand the underlying principles and explore alternative solutions. For example, if you’re looking for a specific data structure implementation:
// Instead of just searching for "Python linked list implementation"
// which might give you the most common or 'accepted' answer,
// try to explore variations or theoretical discussions.
// Example of a basic linked list node
class Node:
def __init__(self, data):
self.data = data
self.next = None
// Example of a simple linked list
class LinkedList:
def __init__(self):
self.head = None
def append(self, data):
new_node = Node(data)
if not self.head:
self.head = new_node
return
last_node = self.head
while last_node.next:
last_node = last_node.next
last_node.next = new_node
// To actively seek diverse knowledge,
// you might then search for "doubly linked list advantages,"
// "skip list use cases," or "comparison of array vs linked list performance."
// This pushes beyond the immediate, predicted solution.
Another practical example could be curating your social media feeds. Many platforms offer tools to mute keywords or unfollow without unfriending. Use them. If a particular topic or type of content is dominating your feed, and you feel it’s narrowing your perspective, actively prune it. It’s a manual override for the algorithm’s assumptions about your preferences.
// On Twitter (now X), for example, you can mute words or hashtags.
// Go to Settings and Privacy -> Privacy and Safety -> Mute and Block -> Muted words.
// Add terms that are dominating your feed negatively or excessively.
// This is a direct intervention to shape your information environment.
// Similarly, on Instagram, you can "mute" specific accounts (posts or stories)
// without unfollowing them, which reduces their algorithmic weight in your feed.
// Tap the three dots next to a post, then "Mute."
These aren’t glamorous hacks, but they are concrete actions that assert your preference over the platform’s.
The Long Game of Agency
This isn’t about blaming the developers or the companies. Their goal, from a business perspective, is often to maximize engagement and user satisfaction, and predictive systems are incredibly effective at that. The challenge, then, falls to us, the users, to understand how these systems work and to develop strategies to ensure they serve our broader interests, rather than simply our immediate, algorithmically-suggested desires.
Our agency isn’t a fixed, immutable thing. It’s something we exercise, cultivate, and sometimes, fight for. In the age of constant prediction, asserting our right to surprise ourselves, to explore the unknown, and to make choices that aren’t pre-ordained by data points, is more important than ever. It’s how we keep our intellectual curiosity alive, how we remain open to new ideas, and how we continue to evolve as thinking, choosing beings.
Actionable Takeaways:
- Periodically break your patterns: Actively seek out content, products, or information that is outside your usual consumption habits.
- Question the source: Before accepting a recommendation, consider why it’s being shown to you and whose interests it serves.
- Set intentions: Approach your online interactions with a clear purpose to avoid passive consumption.
- use manual controls: Use platform features like muting keywords or curating feeds to override algorithmic assumptions.
- Explore non-personalized avenues: Seek out forums, blogs, and other communities driven by human curation and shared interest.
Stay curious, stay unpredictable, and keep pushing back against the easy path. Until next time.
Related Articles
🕒 Last updated: · Originally published: March 24, 2026