AI chatbots such as ChatGPT are increasingly being used by the global public—particularly younger users—to access news, according to the Reuters Institute for the Study of Journalism’s annual Digital News Report, released on Tuesday.
The 2025 edition of the widely regarded survey, conducted by YouGov across 48 countries with a sample of 97,000 respondents, documents for the first time a measurable shift towards using AI-driven chatbots to retrieve news headlines and updates. Although only seven percent of the general population reported doing so, the figure rose to 12 percent among under-35s and 15 percent among those under the age of 25.
“This is the first time we have seen significant numbers of people using chatbots for news,” wrote Mitali Mukherjee, the director of the Reuters Institute, which is affiliated with the University of Oxford.
OpenAI’s ChatGPT was identified as the most widely used chatbot for news consumption, followed by Google’s Gemini and Meta’s Llama. Respondents cited the ability of these tools to provide relevant and personalised updates as a primary reason for their usage. In addition to direct news retrieval, a larger share of users said they employed AI for tasks such as summarising articles (27 percent), translating content (24 percent), and generating recommendations (21 percent). Nearly one in five reported asking AI questions about current events.
However, the report highlights persistent concerns about trust and transparency. Many respondents expressed scepticism about the reliability of AI-generated news, fearing that it could render journalism less accurate and harder to verify. One of the core issues stems from the nature of large language models (LLMs), which are not explicitly programmed but trained on extensive datasets, including publicly available text, images, and videos—often sourced from media outlets.
These models can generate fluent and seemingly authoritative responses to natural-language prompts, but are known to occasionally produce so-called “hallucinations”—fabricated content that appears plausible but lacks factual basis. This limitation has prompted both commercial collaboration and legal conflict between AI developers and news organisations.
While some outlets have opted to partner with AI firms—Agence France-Presse (AFP), for instance, has granted French company Mistral access to decades of archival news—others have taken legal action. The New York Times is currently pursuing litigation against OpenAI, alleging unauthorised use of its copyrighted material during model training.
The broader media landscape outlined in the report continues to shift away from legacy platforms. Traditional outlets—television, radio, print newspapers, and even dedicated news websites—are losing audience share to digital platforms and social media. Among 18 to 24-year-olds, nearly half now cite platforms such as TikTok as their primary source of news, with this trend most pronounced in countries including India, Brazil, Indonesia, and Thailand.
Despite polarisation and ongoing concerns over moderation and content quality, the social media platform X (formerly Twitter) remains a prominent channel for news consumption. The Reuters Institute found that usage of X for news had increased notably in several countries, including the United States, where 23 percent of respondents said they relied on it—an eight percentage point rise from 2024. Usage also increased in Australia and Poland.
The report notes that X has attracted a more right-leaning user base since being acquired by Elon Musk. “Many more right-leaning people, notably young men, have flocked to the network, while some progressive audiences have left or are using it less frequently,” the authors noted.
In contrast, newer alternatives such as Threads, Bluesky, and Mastodon have yet to gain meaningful traction. Each was found to have a global reach of two percent or less when it came to news consumption.
Overall, the Reuters Institute’s findings underscore a transitional moment in the global information ecosystem. While the reach of traditional media continues to erode, a generational shift towards AI-enhanced tools and video-based platforms appears to be accelerating. The report highlights both opportunities for innovation in news delivery and significant challenges concerning the accuracy, ethics, and economic sustainability of journalism in the AI age.
Read also:
Latest AI Models Show Greater Power – But Also More Frequent Errors, Studies Find
