Restoring the Digital Ecosystem: Why AI Should Help Us Explore Again
The Internet is Shrinking, and AI Might be the Reason
Over the past few years, the internet has started to feel smaller.
Not in size, but in texture. There is less nuance, more familiarity, and ubiquity of thought for those siloed in similar social graphs.
Where we once explored blogs, forums, and conducted late-night rabbit hole deep dives that rewarded curiosity, we now ask a single question and receive a single answer. Don’t get me wrong, it’s efficient and clean, but has it removed something unique to humanity?
AI has made searching beautifully simple. But simplicity always has a trade-off.
From Ecosystem to Extraction
The internet was once a living ecosystem, full of diverse voices and viewpoints, unpredictable paths, and unexpected discoveries.
However, today, with how AI systems consolidate how we find information, the internet risks becoming a resource mine rather than a diverse ecosystem of views and counterviews.
Content has quickly become training data.
Sources have become citations, but not information to be read directly
The open web has faded into the background behind a polished and synthesized response of mundanity.
While we may gain clarity, we lose texture. We’re able to learn about the What, but not Who or Why.
Environmental researchers have long warned that when natural resource systems are mined without renewal, the damage can be profound. A 2021 paper published by The Extractive Industries and Society titled Understanding the impacts of mining on ecosystem services through a systematic review highlighted that active mining operations often cause negative harm to ecosystem services, thus weakening the very services that sustain human wellbeing. While the study focuses on the impact of physical mining on communities, a similar dynamic now threatens the digital world. As AI systems continue to extract knowledge from millions of sources without reinvesting in their vitality, the internet risks shifting from a living ecosystem into a finite resource mine.
Incentives Are the Invisible Architecture of Truth
When we live in a digital world built from extraction, the real question isn’t just how AI gathers knowledge. It becomes “What is it paid to say to us?”
If advertising becomes the dominant business model, it will inherit the same misaligned incentives that warped social media. Specifically, engagement over accuracy, persuasion over truth, and invasion of privacy in the pursuit of highly targeted and manipulative content. As highlighted in The Journal of Economic Perspectives paper The Online Advertising Industry: Economics, Evolution, and Privacy, online advertising was a “gale of creative destruction”, but have we effectively addressed the negative externalities from the initial model? Many academics and papers would argue we’ve not, and hardly any legislation has been passed to change this perspective.
Advertising doesn’t corrupt directly, but it tilts everything it touches. When the persuasive power of language meets generative AI, the line between simple guidance and manipulation will become very blurry, very quickly.
So perhaps the modern question isn’t “Will AI lie?” It’s “Who profits when it does?”
Curated Intelligence or Collective Discovery?
It’s clear we’re standing at a crossroads of sort.
One path will lead us to curated intelligence, where AI mediates most human knowledge and gives us quick, context-free answers.
While the other path leads us to collective discovery. AI could be seen as a companion that helps us navigate the messy web of ideas, guiding us through the internet rather than around it.
Both paths have something to offer:
The first is faster, safer, and more censored, and likely more profitable in the short term.
The second is slower, more chaotic, and uncontrollable. It is human in its approach but partnered and supported by machines. The perceived value and ability to profit from this model is unknown.
Ultimately, our choice will determine whether the next generation learns to question or simply consume based on an AI’s response.
Finding Balance
A healthy digital ecosystem will depend on a few key principles:
Transparency: Show how and why information is being surfaced to the individual or user.
Agency: Allow people to switch between models, summary responses, or full explorations of information.
Open-Source: Support diversity in the creation of content and ensure open data access to drive accountability, open dialogue, and critique.
Align Incentives: Reward truth-seeking behavior and curiosity, the willingness to acknowledge and correct ill-informed perspectives or opinions, and reduce the reliance on attention.
If AI becomes the only lens through which we see the web, we risk forgetting how to search, verify, and think critically. This is the very foundation on which the internet was established, and by extension, the meaning of being human. We should hold this dear to our hearts and minds, and consciously combat a quiet and gradual decline in one’s ability to think for oneself.
The Real Measure of Progress
True progress isn’t about replacing human discovery with machine efficiency. It’s about making curiosity easier to act on.
Because when we stop exploring, the internet doesn’t just shrink. We Do Too.
Image was generated using ChatGPT Image Generator 4o