To Their Credit

Journalism

Nikkei panel puts trust at the centre of journalism’s AI search shift

Publishers face falling search referrals and a new contest to be visible, credible and transparent in AI-generated answers.

Ledger: DisputedNo outcome yet

News leaders meeting in Tokyo for Nikkei’s 150th-anniversary programme said publishers’ ability to retain audience trust, and increasingly to be recognised as trustworthy by artificial-intelligence systems, will be central as AI changes how people discover and consume news.

The discussion took place on Nov. 18 during a session titled “Future of News - Management.” Participants also stressed transparency as a defence against misinformation. Nikkei’s programme cast the task for publishers as responding to social media and AI while preserving public confidence in quality journalism.

The event linked editorial credibility to the economics of independent reporting. Its programme said stronger, more resilient media businesses were essential to sustaining independent journalism at a time of rapid technological and informational change.

The emphasis echoed concerns raised by The Guardian at a Seoul journalism conference earlier this month, as we reported Nov. 14, that AI-generated answers could weaken publishers’ direct connections with readers.

The traffic pressure is already measurable. Chartbeat data cited by the Reuters Institute showed Google Search referrals to hundreds of publishers fell 40% between July 2025 and July 2026, twice the decline recorded in the preceding 12 months. Readers are increasingly receiving answers through Google AI Overviews and chatbots rather than clicking through conventional results.

That does not mean chatbot news use is yet universal. Weekly generative-AI use across six countries rose from 18% in 2024 to 34% in 2025, but only 10% of the global sample in the Digital News Report 2026 said they used chatbots for news. Use was higher among younger people.

Publishers are responding by pursuing more direct audience relationships and monitoring whether they appear in AI answers from products including ChatGPT, Perplexity and Google AI Overviews. AI systems retrieve and synthesise material from selected sources, but their citation patterns differ considerably by platform and are often concentrated among a small number of dominant outlets.

Relevance, recency, statistics and quotations can improve the likelihood of appearing in AI citations, although there is no mature equivalent of search-engine optimisation for generative AI. Established brands, including Reuters, the BBC, The New York Times and The Guardian, feature frequently, while smaller publishers may have an opening in specialised or local information that larger outlets do not supply.

The incentives create a difficult choice over access. By the end of 2023, 48% of major news sites in 10 countries were blocking OpenAI crawlers. Yet blocking retrieval bots can also reduce the chance that a publisher’s reporting is included in AI-generated answers.

Nikkei has separately sought to address provenance and manipulation risks. It joined the Content Authenticity Initiative in July 2024, saying fake images and videos using generative AI posed a major threat to media and that participation could increase transparency around the photographs and videos it publishes.

See also

More from Journalism

All Journalism news
  1. Vietnam journalism seminar puts human judgment at centre of AI trainingLedger: Disputed
  2. Mississippi workshop showed student journalists how AI can surface local story leadsLedger: Adopted
  3. India’s Press Council updates journalism norms for an AI and deepfake eraLedger: Resisted