Gresham's Law in the Age of AI
Gresham's Law in the Age of AI
By Jamie Watters | April 2026
In the 16th century, Sir Thomas Gresham observed something counterintuitive about money: when two forms of currency circulate simultaneously, the inferior one drives out the superior.
"Bad money drives out good."
The mechanism is simple. If you have a genuine gold coin and a debased coin of the same face value, you spend the fake and hoard the real. Multiply that behaviour across an economy and the debased currency floods the market while the genuine article disappears from circulation.
Gresham's Law was about coins. But the principle applies anywhere value can be counterfeited — and right now, it is playing out in real time across the entire digital content ecosystem.
The Debased Currency of the Internet
Generative AI has made it possible to produce content at near-zero cost. Blog posts, LinkedIn updates, YouTube scripts, product landing pages — all can be generated in seconds. The barrier to entry has not just been lowered. It has been obliterated.
This is the "bad money" of the internet era: cheap, low-effort AI content produced in volume with the sole purpose of gaming recommendation algorithms, capturing attention, and extracting value before the audience realises there is nothing underneath.
It is not inherently dishonest. A single-shot prompt producing generic text about "5 productivity hacks" is not a lie. But it contains no original thought, no lived experience, no unique perspective. It is a recompilation of what already existed — the AI content machine is, as Richard Cole of the University of Bristol described it, "fundamentally a recompilation engine of what we have already done."
When this debased content floods the market, it creates a problem that mirrors Gresham's original observation: genuine insight gets crowded out. The person who spent three weeks running an experiment and writing up their real findings competes for the same attention as the person who generated 50 posts in an afternoon. On a surface level, they look identical.
The Short-Term Illusion of Success
The snake oil salesmen of the AI content era have found a real seam to mine, at least temporarily.
Volume arbitrage works. Flood enough channels with enough content and some of it converts. Fake it with enough confidence and some people will believe you have authority you have never earned. The economics are seductive: the cost of production is nearly zero, and even a tiny conversion rate generates positive returns.
But this model depends on a gullible audience that cannot distinguish genuine coins from debased ones.
That audience is shrinking.
Consumer preference for AI creator content has already collapsed from 60% to 26% in three years. Researchers are documenting "moral disgust" responses to AI-authored emotional content. Platforms are racing to build provenance tools. The EU AI Act mandates transparency requirements. Instagram's CEO released a memo calling for AI labelling and ranking penalties for non-original content.
The economic model of the AI content grifter depends on the gap between "looks credible" and "is credible" remaining invisible. That gap is closing faster than most of them realise.
Why the Genuine Article Appreciates
Here is what Gresham's Law also implies, and what most people miss: the good money does not disappear. It gets hoarded. It becomes scarce. And scarce genuine value, in a market flooded with counterfeits, becomes more valuable, not less.
The same dynamic is playing out in content. As the volume of AI slop increases, the signal-to-noise problem worsens for everyone. Finding genuinely useful, original, experience-backed content becomes harder. That difficulty makes it more valuable when you find it.
Google's E-E-A-T framework — Experience, Expertise, Authoritativeness, Trustworthiness — is the operational expression of this principle. Search engines are building systems to distinguish genuine coins from debased ones. AI discovery platforms are doing the same. A landmark Nature study showed that AI models trained on AI-generated data undergo "model collapse" — they literally need original human content to remain useful.
The architecture of the internet is being rebuilt around provenance and proof. The people investing in genuine expertise, transparent track records, and first-hand experience are accumulating the asset that is about to appreciate significantly.
The ultimate defence against Gresham's Law — in both economics and content — is AI literacy. When enough people can spot a debased coin, the bad money loses its power. We are approaching that moment faster than the slop machine operators appreciate.
Where I Am Betting
I build in public. Real metrics. Real failures. Real lessons from actually shipping products.
That transparency is not just a philosophy. It is a strategic asset — the unforgeable cryptographic signature of genuine experience. No AI can fabricate a two-year public trail of building, failing, learning, and iterating. No language model has skin in the game.
The bad money is circulating. But the good money is being hoarded by the people who built it the hard way.
I know which I am producing.
This is part of the Fool's Gold series. Read the pillar article: The Road to Hell is Paved with AI-Generated Fool's Gold
Building in public at jamiewatters.work