A global survey conducted in 2025 by the Reuters Institute and the University of Oxford has revealed a significant surge in public awareness and utilization of generative artificial intelligence. Weekly usage of tools like ChatGPT has nearly doubled, jumping from 18% in 2024 to an impressive 34% in 2025.
The study, which included participants from Argentina, Denmark, France, Japan, the United Kingdom, and the United States, investigated the integration of generative AI across various domains, including news, search engines, and broader societal contexts. ChatGPT emerged as the most popular and recognized tool, with 22% of respondents reporting using it weekly.
The level of awareness surrounding generative AI tools has substantially increased across all surveyed countries. An impressive 90% of participants acknowledged being aware of at least one tool, marking a notable increase from 78% recorded the previous year. Younger demographics, particularly those aged between 18 and 24, demonstrated the highest adoption rates.
Information-seeking has surpassed content creation as the primary application for generative AI. Weekly usage for information retrieval has doubled to 24%, while media creation saw an increase to 21%. Despite these gains, the adoption of AI for news-related purposes remained limited, with only 6% of respondents using AI systems to access news on a weekly basis. Argentina and the United States recorded the highest usage rates for news consumption through AI.
The survey revealed mixed feelings about the trustworthiness of AI-generated search results. "While 54% reported seeing AI-generated answers in searches, only half of those said they trusted the information," according to the report. Click-through rates on source links remained low, and a significant number of users preferred to verify AI-generated responses using traditional sources, especially for topics like health and politics.
The study also highlighted the perception that AI is already prevalent across various sectors. Over half of the respondents believed that AI is commonly used in news media, with even higher figures for search and social media companies. However, optimism regarding the impact of AI varied depending on the sector. While respondents were more positive about AI in healthcare, science, and search engines, they expressed more reservations about its use in news media, government, and politics.
In the realm of journalism, a significant disparity exists between comfort levels with human-produced and AI-produced content. Only 12% of respondents expressed comfort with fully AI-generated news, compared to 62% who preferred human-created content. Acceptance increased to 43% when a human oversaw the process with AI assistance.
People were more receptive to the use of AI in back-end tasks, such as grammar correction and translation, compared to front-facing applications like AI-generated images or artificial presenters. The survey showed that only one-third of respondents believed that journalists regularly verify AI-generated content before publication, and 43% anticipated substantial differences in responsible AI usage across various news outlets.
Despite the majority of people consuming news daily, only 19% reported encountering AI labeling on news stories regularly. AI summaries and chatbots were the most commonly recognized features, but 60% of respondents stated that they did not routinely encounter AI features on news platforms. The report also identified variations in perception across different countries. Japan and Argentina generally displayed greater optimism regarding the societal impact of AI, while the United Kingdom and the United States showed greater skepticism.
These findings are part of the Reuters Institute’s ongoing research into public attitudes toward AI in journalism and society, which is supported by the National Center for AI in Society and the Thomson Reuters Foundation.