Who Will Monetize Truth? A Thesis For the Future of Information
The news industry isn't declining. It's being repriced around a single distinction: some companies sell awareness of what happened, others sell the ability to act on it first. The first charges $10 a month. The second charges $30,000 a year.
THESIS · 01 MAR 2026 · FRANCESCO MARCONI
For twenty years, the story about media has been a eulogy. Revenue down. Newsrooms shrinking. Print collapsing. The story is accurate. It is also useless, because it describes the symptoms of a phase transition without identifying the transition itself.
The news industry isn't gradually declining. It's being repriced around a single distinction: some companies sell awareness of things that happened. Others sell the ability to act on information before everyone else can. Same raw material, the same reporters, the same filings, the same data. Opposite economics.
Content informs awareness. Intelligence informs a decision.
$52B
Revenue from companies selling intelligence, from Bloomberg, S&P Global, Moody's and peers, at 35%+ margins
$21B
Revenue from the entire U.S. newspaper industry, at margins near zero
275:1
What an intelligence terminal costs versus a consumer news subscription
The value didn't disappear from journalism. It migrated, from the content layer to the intelligence layer. When content becomes abundant, it loses pricing power. When everything is noise, the scarce resource is the ability to detect signal. Value always migrates to the scarce layer.
Three species, one ecosystem
Plot every major information company on two axes: what share of revenue comes from data and intelligence products, and how much revenue it generates per employee. Two clusters emerge. They are two different businesses. The companies in the middle, Dow Jones, the Financial Times, Schibsted, Hearst, discovered they contained both: a content operation and an intelligence operation, sharing the same newsroom but serving different buyers.
01
The Intelligence Business
Sells the ability to act on information before others can. $5K to $32K per year.
02
The Attention Aggregator
Sells awareness of things that happened. Traffic collapsing 50%.
03
The Public Good
Local accountability, investigative work. Needs a different funding model.
Most media institutions don't know which species they are. The classification is the whole game, and it determines what happens to those who get it wrong.
Fifteen questions about where value is moving
The full thesis is structured as fifteen questions about where value, talent and power are moving in the information economy. The argument, in compressed form:
Q01
Where does value accrue?
Content is free. Intelligence is not.
Q02
Who wins and loses among media institutions?
Three species. Only one has pricing power.
Q03
What happens when AI agents become the primary consumers of news?
51% of web traffic is now non-human. The audience has already left.
Q04
Is information advantage the new alpha?
There's a gap between when a signal appears in public data and when it becomes a story. That gap can now be priced.
Q05
What does the intelligence layer look like outside finance?
Bloomberg built the terminal for financial data. No one has built the equivalent for clinical trials, energy or cybersecurity.
Q06
If AI can hallucinate citations, what happens to the concept of the record?
The factual record can now be fabricated with perfect confidence.
Q07
Are prediction markets the next major consumer of structured information?
$40B in volume. They price the fact itself. The infrastructure to resolve those facts doesn't exist yet.
Q08
Is trust mispriced?
The same FDA filing is worth $0, $50 or $50,000 depending on the decision it informs.
Q09
If this becomes a data game, what's the trade?
AI companies spent $200B on infrastructure. They paid content producers 1.5% of that.
Q10
What happens when AI runs out of journalism to train on?
AI is getting smarter about a world that no longer exists.
Q11
Can wire services capture the value they produce?
The ones who structure their data for machines win. The ones who keep shipping prose won't.
Q12
What can't be automated?
Observation is getting cheaper. Interpretation is not.
Q13
If AI writes all the articles, what is the newsroom actually for?
The article is the exhaust product. The detection pipeline is the real output.
Q14
What happens when $1 of AI replaces $33 of freelance labor?
The bottom fell out. But the top is higher than it's ever been.
Q15
What happens to the New York Times? What happens to Google?
Both bets are working. For now.
THE FULL THESIS
The argument runs deeper than the summary.
Fifteen questions, the full repricing dataset, and where the intelligence layer opens next.
