Reuters: how the public in six countries think about generative AI in journalism
Generative artificial intelligence, once considered futuristic, is now becoming part of newsroom workflows, with major organizations like the Associated Press and Bloomberg incorporating it into their news production. But how does the public feel about this?
A recent study, “What Does the Public in Six Countries Think of Generative AI in News?” from the Reuters Institute sheds light on public perceptions of generative AI in journalism across six countries. While there is curiosity about AI’s role in the news, many people still prefer the traditional, human approach to reporting.
Comfort levels with AI in news
The study found that most people are more comfortable with news produced by human journalists. While some newsrooms, like The Washington Post and Bloomberg, have started using AI to write or assist in creating stories, the general public remains cautious. Interestingly, the study reveals that 62% of younger audiences, particularly those aged 18 to 24, are more open to the idea of AI-generated news compared to older audiences. In countries like the United States and Argentina, acceptance of AI in journalism is a bit higher, but overall, skepticism is the dominant feeling.
Different reactions by type of news
Opinions often depend on the type of news being covered. People tend to be more open to AI-generated stories for lighter, less serious topics such as fashion or sports reporting. However, when it comes to hard news, like politics or international affairs, most people prefer that humans handle the reporting. This difference in comfort levels suggests that people may be more willing to accept AI in less critical areas of news.
Trust, quality, and value concerns
One of the biggest barriers AI faces in journalism is public trust. Many respondents expressed concerns about the reliability and accuracy of AI-generated news. People worry that AI might not be able to provide the same level of depth, transparency, or investigative rigor as human journalists. Trust is a significant issue, with 47% of people saying they wouldn’t be willing to pay for news created by AI. Even though AI can produce news quickly and cost-effectively, the perception of lower trustworthiness is something news organizations will have to address.
Where AI fits best in newsrooms
Although respondents were hesitant about AI taking on a major role in news creation, there are areas where they were more accepting of AI’s involvement. Many felt comfortable with AI handling tasks behind the scenes, like editing articles or analyzing data. AI can improve efficiency without directly influencing the news content in these areas. However, when it comes to more visible roles, such as AI-generated anchors or fully automated news broadcasts, public discomfort increases significantly. Around 95% of respondents said that if AI is used in any part of news production, it should always be clearly labeled.
Awareness of AI tools
While many are aware of generative AI tools like ChatGPT, not everyone is using them regularly. The study found that younger and more educated individuals tend to have a more positive outlook on AI in general, but regular usage is still relatively low. This lack of everyday interaction with AI may be contributing to the skepticism surrounding its role in journalism. Familiarity with AI technology appears to shape people’s views, with those who are more aware or experienced with AI tending to be more open to its use in news production.