We were promised an internet of limitless access — a place where the curious could dig, cross-check, and learn from the messy abundance of perspectives. What we got instead is a slow-motion coup, where convenience overthrows accuracy and nobody even fights back.
The battle lines aren’t drawn in smoke-filled rooms or parliamentary chambers; they’re coded into algorithms, UX flows, and the default settings of AI systems. “Good enough” is now the gold standard — not because it’s better, but because it’s faster.
From Precision to Plausibility
Platforms have quietly shifted their objectives. Once, a search engine’s worth was measured by the precision of its results. Now, plausibility is enough. Large Language Models (LLMs) like GPT, Claude, and Gemini don’t claim to be right; they claim to be useful. Fact-checking layers on platforms aren’t shields for truth — they’re reputational insulation for the platform itself.
The irony? The more we rely on AI to give us answers, the more we blur the line between accurate and acceptable. A headline that “sounds true” is good enough to share. A chatbot’s authoritative tone is enough to believe.
The Role of the Lazy Scroll
But this isn’t just an AI problem — it’s a user problem. We reward speed over substance. We’ll skip the 12-minute explainer for a 30-second recap, then mistake our recognition of the summary for actual understanding. LinkedIn thrives on this dynamic — entire industries reduced to bullet-point “frameworks” and viral “five hacks” threads. The platform has gamified shallowness, and its reward system actively punishes those who write beyond skim length.
The result is an attention economy where “what travels” beats “what’s true” every time.
AI as an Accelerant
In communication theory, Marshall McLuhan noted that the medium shapes the message. AI isn’t just a medium — it’s an accelerant. It takes the pre-existing tilt toward speed, compresses it further, and automates the spread. LLM outputs aren’t encyclopedias; they’re acceptable approximations, delivered with enough confidence to discourage deeper digging.
The danger isn’t that AI is wrong — it’s that AI makes “good enough” feel like “good enough forever.” When the first answer feels final, the incentive to keep asking disappears.
The Erosion of Epistemic Patience
Once, knowing something took effort. You had to read, cross-reference, and hold contradictory information in your head long enough to sort it. That cognitive endurance — epistemic patience — is eroding. Platforms don’t just accommodate this loss; they exploit it.
The “convenience coup” works because it’s not an imposition; it’s a seduction. We choose it. We opt into the easier path. And every time we do, we make it harder to imagine why accuracy was worth the trouble in the first place.
What’s at Stake
If communication becomes permanently optimized for plausibility over truth, the consequences are not just individual but systemic. Policies will be shaped by what’s perceived to be right rather than what is right. Brands will measure “trust” in likes rather than in evidence. And the role of media literacy will shift from teaching people how to find the truth to teaching them how to survive without it.
The coup isn’t coming. It’s already happened. And most of us clicked “Accept All” without reading the terms.
