The Counterfeit Baseline: Why Disinformation Is Redundant

Media has never been primarily a truth-delivery mechanism. AI didn’t corrupt the ecosystem — it just made the counterfeiting cheaper.

The first thing Gutenberg chose to mass-produce was not a dictionary. Not a scientific treatise. Not a census.

It was a text asserting its own divine authority — a document whose truth claim was structured to resist external verification. “This is true because it says so, and it says so because it’s true.” That’s not a flaw in the medium’s debut performance. It’s the medium demonstrating its core capability on launch day.

We’ve been treating this as a historical curiosity. It’s actually the whole template.

Before Print, There Was the Shaman

Writing itself — cuneiform, Linear B, early Egyptian hieroglyphics — didn’t emerge primarily to record knowledge. It emerged to record debts and royal proclamations. The oldest written artifacts are tax records and monuments to rulers. Accurate accounting of reality was a secondary use case from the beginning.

Oral tradition preceded writing, and it was no more honest. The shaman who controlled the story of where lightning came from controlled community behavior during storms, and by extension a great deal else. The information was almost entirely constructed. The construction was the point.

Each medium since has followed the same sequence: a new transmission technology appears, optimists imagine it will spread truth, and within a generation it has been colonized by whoever most urgently needs to construct a shared reality. The printing press gave us the Reformation and the Index Librorum Prohibitorum simultaneously — competing constructed realities, with power determining which one won.

The Compression Is the Story

What changes across media transitions isn’t the function. It’s the speed.

Gutenberg’s press took roughly fifty years to become a serious propaganda instrument — partly because literacy itself had to diffuse first. Radio weaponized in about fifteen. Television in roughly the same. Social media took perhaps five years between mass adoption and Cambridge Analytica. The compression isn’t coincidental.

Each new medium arrives into an existing infrastructure of influence: an advertising industry, a political consulting class, an attention economy with established monetization channels. The medium doesn’t have to build that apparatus from scratch. It plugs in.

AI plugged in on day one. There was no latency period, no naive early era of honest deployment. The tools for constructing reality at scale were already there, waiting.

The Byte Deflation Problem

There’s a subtler dynamic underneath the obvious one. The leverage of a single unit of information — call it a byte — has been falling for decades, and AI has pushed it close to zero.

When Gutenberg printed a page, scarcity conferred authority. The fact that something was printed implied someone had paid to print it. That credibility premium has been entirely inflated away. A byte today carries essentially no prior probability of being true, important, or intentionally authored by a human being.

Human cognitive bandwidth hasn’t changed. Attention is a fixed resource. The leverage of any given byte is roughly attention captured divided by bytes produced. The numerator is flat. The denominator is now growing faster than at any point in history.

Counterfeiting works until the counterfeit supply exceeds the economy’s ability to absorb it. At sufficient volume, the currency itself becomes worthless — not this bill or that bill, but the category. We are approaching that threshold for information. Not because any particular piece of content is dishonest, but because the ratio of signal to noise has collapsed past the point where the distinction is operationally useful.

The Incoherence of “Disinformation”

This is where the standard framing goes wrong.

“Disinformation” implies a baseline of formation — true information against which the false can be measured. But if media has functioned primarily as a consensus-manufacturing technology since before writing was invented, the baseline is not truth. It is negotiated reality, with power determining whose negotiation wins.

The scientific paper comes closest to genuine exception — an institution explicitly designed to resist the consensus-manufacturing impulse. But publication bias, the replication crisis, and the social dynamics of peer review suggest the impulse reasserts itself even there. The institution built to resist it has been partially captured by it.

This doesn’t mean lies don’t matter. It means the question “is this true?” may be less useful than “whose constructed reality does this serve, and what are the mechanisms by which it is being enforced?”

That’s a harder question. It’s also the right one.

What AI Actually Changes

AI doesn’t introduce manipulation into the information ecosystem. It removes the last meaningful constraint on its scale.

Previously, constructing a convincing false reality required human time — writers, editors, designers, distributors. That time cost imposed a natural ceiling on the volume of manipulation any single actor could produce. The ceiling was already falling fast with social media’s algorithmic amplification. AI removes it entirely.

More significantly, AI shifts manipulation from broadcast to individuated. Every previous medium sent one message to many recipients. The influence was statistical and approximate — you could move populations but not individuals. Personalized manipulation at broadcast scale is structurally different. It isn’t an incremental improvement in the Joules-to-behavior-change pipeline. It’s a phase transition.

The regulatory apparatus built for broadcast — equal time rules, libel law, spectrum licensing — has no obvious analog for individuated influence at scale. Which is one reason the attempts to regulate AI information effects keep reaching for the wrong tools.

The Scarce Resource

If bytes are inflating toward worthlessness and individual influence is now cheap to manufacture at scale, what actually becomes valuable?

The trusted filter. The curation layer that tells you which bytes deserve your attention. Not the information itself, but the credible claim about which information matters.

That’s what everyone from search engines to social platforms to AI assistants is actually selling. And it’s what every government, every media institution, and every major technology company is quietly fighting over.

The medium that wins isn’t the one that produces the most bytes. It never was.