Picture this: democracy, science, and the translation industry walk into a bar. Democracy orders facts, science asks for nuance, and the translators — already behind the counter with a stack of mangled receipts — quietly say: “We’ll take whatever readable English you’ve got.”
And of course, if those receipts are illegible, everyone ends up footing the bill.
In other words, while many people point to AI as the storm on the horizon, the real leak in the roof is something far more mundane — and far more pervasive: the decline of careful reading, critical attention, and high-quality written communication. AI may amplify noise, but post-literacy is the erosion of the plumbing itself. With that in mind, let’s walk through why the real danger starts long before the algorithm kicks in.
1) The data: reading skills are slipping
To begin with, major international assessments continue to show that adult literacy and numeracy have either stagnated or declined in many high-income countries. In fact, depending on the dataset, roughly one in five adults performs at a level comparable to a primary-school pupil on basic literacy tests. As a result, the foundation on which democratic debate, scientific understanding, and professional communication rest is becoming increasingly shaky.
2) Why that’s bad for democracy
Building on this, democracy relies on shared facts, functional comprehension, and the ability to evaluate competing arguments. Yet as careful reading diminishes, manipulative content and half-truths slip in more easily. Numerous studies show that stronger information literacy improves resilience to misinformation, whereas weaker literacy effectively swings the gates wide open. In short, when citizens skim instead of scrutinize, democracy becomes easier to derail.
3) Why that’s bad for science
Furthermore, science doesn’t thrive on vibes; it thrives on comprehension, skepticism, and the willingness to follow complex chains of evidence. If readers lack the patience to parse methods, limitations, or uncertainty, nuance turns into noise while oversimplified narratives harden into memes. Consequently, the public conversation shifts from evidence to emotional certainty — never a good trade.
4) Why that’s bad for translation
And then there’s the translators — the quiet custodians of clarity. Translators don’t work in a vacuum; they work from source texts. When those texts are sloppy, vague, or incoherent, translators face an impossible triad: guess, leave blanks, or produce a hyper-literal mess. In this way, declining literacy upstream inevitably leads to degraded communication downstream. Bad input = bad output — even with the best linguists in the world.
5) Where AI fits in
At this point you might ask: doesn’t AI pose a threat too? Absolutely — but in a different way. AI can scale misinformation at unprecedented speed. However, AI doesn’t create the underlying vulnerability; it merely exploits a readership that doesn’t check, skims instead of parsing, and prefers emotional resonance over evidentiary weight. Put simply, AI is the megaphone; post-literacy is the dry kindling that makes the megaphone dangerous.
6) What to do
Because the problem is upstream, the solutions must be too. That means:
- Reinforcing deep reading as a social habit.
- Investing in media and information literacy.
- Improving source-text quality in business, education and government writing.
- Treating AI as a tool requiring guardrails, oversight and human competence — not as a replacement for literacy.
Final bite
If democracy is a potluck, then literacy is the recipe card. Post-literacy is when half the guests arrive with mystery casseroles and no labels. AI, meanwhile, is the loudspeaker blasting the “taste this!” jingle. Want to save the meal?
Teach people to read the recipe, label every dish, and — above all — don’t let the loudspeaker hand out spoons engraved with “trust me.”