← All media signals
Media signal

The Human Filter

As generated content floods every feed, platforms are rebuilding editorial surfaces where a real person, not an algorithm, decides what gets shown first.

What is it?

As generated content fills every feed, platforms are rebuilding editorial surfaces where a named person, not a ranking model, decides what gets shown first. Newsrooms hold the line on human-written copy while automating everything around it. Whether that discipline scales past one flagship newsroom stays open.

Why now?

Automation was supposed to make selection cheaper everywhere, on every feed. Instead the flood of synthetic material made a verified, named human choice scarcer, and scarcity is what carries a price in a crowded market. The premium runs in reverse of what automation usually promises to deliver.

Who’s building it?

Newsrooms and platforms reintroducing editorial roles they had spent years trying to automate away, with verification vendors forming underneath them.

Newsroom editorsPlatform curatorsVerification vendorsAI summarisation toolsTrust-and-safety teams
The full read
Why now, who is building it, where the market goes, and what stays unsettled.
Get weekly →
01 · The market
Who’s in it
Builders — newsrooms and platforms rebuilding named, human-led editorial surfaces.
Buyers — publishers and platforms weighing editorial cost against algorithmic scale, plus trust-seeking audiences.
Gatekeepers — the newsrooms whose editorial standards others cite as the model.
What’s at stake
Publishers weigh editorial cost against the scale advantage automation offers — trust and margin pull against each other.
A named editor’s judgment is hard to fake convincingly.
Verification becomes its own billable line, not a background cost.
What’s unsettled
Whether audiences actually pay more for a human byline.
Verification as a product line, or just marketing language.
Whether smaller newsrooms can afford to build the human layer at all.
02 · How the demand matureda progression, not a timeline
Saturation
Feeds fill with synthetic material until every surface starts to feel interchangeable.
Signal
A disclosed human selection becomes a mark that readers begin to notice, trust and actively seek out.
Premium
Verified curation earns a price separate from the underlying content itself, not folded into it.
Standard
Disclosure of human involvement becomes an expected baseline, not a differentiator.
03 · Where the market goesthe forks, not a prediction
→Structural response or nostalgia — decided by whether readers keep paying once the novelty of a human byline wears off.
→One flagship newsroom’s practice could set the model every smaller outlet is measured against.
→Verification could stay a marketing claim, or harden into an audited, billable service line.
→Platforms may absorb curation themselves rather than pay newsrooms for it.
Everything here turns on willingness to pay: whether a verified human choice is worth more than a good algorithmic guess, or just feels that way for now.
04 · The Taste Economy deskadjacent markets, moving now
The Taste Broker
Paid subscriptions cross five million readers.
Substack passed 5 million paid subscriptions in March 2025.
Tubefilter
The Taste Graph
Platforms now model taste directly, not just clicks.
Pinterest’s Taste Graph maps people to interests and aesthetics.
TechCrunch
Paid For Taste
Editorial judgement starts commanding a real price.
Trade commentary argues taste was always the business model.
lokol (Substack)
05 · The Wire
01
What automation can’t do.
GeoBarta ↗
02
Verification as its own line.
Media Copilot ↗
03
Publishers split on swapping human newsletter curation for algorithms
Digiday ↗
04
Trust in news from AI chatbots is 20%, versus 37% overall
Reuters Institute for the Study of Journalism ↗
The panel on this signal
They agreed on
–Scarcity, not nostalgia, is what makes a verified human choice valuable in a synthetic feed.
–One newsroom is doing most of the defining right now.
–Verification is heading toward becoming its own priced line.
The behavioral economist disagreed on
–Willingness to pay for editorial curation has tracked its novelty relative to algorithmic alternatives; once human selection becomes routine rather than a differentiator, the premium attributed to it may not hold.
Know more about the method →
✦How this signal was built
01 · Panel
A Delphi panel of 4 AI analysts, each reasoning independently.
→
02 · Debate
They challenge and revise each other’s reads over two rounds.
→
03 · Evidence
Every claim is tied to a cited, linked source on The Wire.
→
04 · Synthesis
A moderator writes it into one read, disagreement and all.