Juni 2024
Tinius Digest
Månedlige rapporter om endringer, trender og utviklinger i mediebransjen.
Om Tinius Digest
Tinius Digest gir deg en oversikt over rapporter om og analyser av utvikling i mediebransjen og publiseres en gang i måneden. Her er våre viktigste funn fra denne måneden.
Del gjerne innholdet med kollegaer og bruk det i møter og presentasjoner.
Innhold
- AI erodes trust and reduces willingness to pay for news
- 3 out of 10 Norwegians are news avoiders
- Norway faces massive potential job automation due to generative AI
- A small group drives most fake news on Twitter
- Loneliness in adolescence is linked to conspiracy beliefs in midlife
- Football fuels 80 per cent of online hate in sports debates
AI erodes trust and reduces willingness to pay for news
Reuters Institute for the Study of Journalism at Oxford University have published a report looking into public perceptions of generative AI in news production across six countries.
Key findings
1
Limited use of generative AI for news
While ChatGPT is widely recognised, only five per cent of respondents across the six countries had used generative AI to access news. Most people use it for creative tasks like generating text or images.
2
Generative AI is expected to reshape news media
Two-thirds of respondents believe generative AI will significantly impact the news industry in the next five years. However, only 12 to 30 percent trust news media to use AI responsibly, highlighting widespread scepticism.
3
AI-generated news seen as less trustworthy
Many participants expect AI-generated news to be cheaper and more up-to-date, but they also predict it will be less trustworthy and transparent. Around 41 per cent of respondents felt that AI-produced news would be less worth paying for compared to news created by human journalists.
4
Comfort with AI in news depends on the topic
Respondents were more comfortable with AI generating news for lighter topics like fashion (+7% net comfort) and sports (+5%) but expressed discomfort with AI handling political news (-33%) and international affairs (-21%).
5
Widespread demand for transparency
The public overwhelmingly supports labelling AI-produced news, with 47 per cent saying it should be disclosed when AI is used to write articles or analyse data. Only five per cent said no disclosure was necessary, reflecting a strong desire for transparency in AI use.
3 out of 10 Norwegians are news avoiders
The 2024 edition of the Reuters Digital News Report explores news consumption, focusing on key trends and concerns, including social media use, news avoidance, and rising scepticism towards AI-generated content. Below are the main findings from the Norwegian findings of the report, produced by the University of Bergen in collaboration with Fritt Ord.
Key findings
1
High daily news consumption
83 per cent of Norwegians check the news at least daily, with local news remaining the most popular. However, younger adults (18–24) have significantly reduced their use of online newspapers and social media for news since 2017. Facebook's role as a news platform has declined, though it remains the most used social media platform for news across age groups.
2
Increased news avoidance
The proportion of Norwegians who actively avoid the news has risen. In 2024, 31 per cent reported often or occasionally avoiding the news. This trend, which was last seen during the COVID-19 pandemic, could be linked to fatigue from ongoing conflicts and political scandals.
3
Scepticism towards AI in news production
Norwegians are generally uncomfortable with the idea of news produced by AI, particularly for serious topics like crime and politics. Only 18 per cent are comfortable with AI-generated news, and scepticism is most vigorous for “hard news”. At the same time, only 37 per cent of respondents felt well-informed about AI.
4
Concerns about fake news remain stable
Around 45 per cent of Norwegians worry about distinguishing between real and fake news online, a figure that has remained the same since 2020. TikTok stands out as the platform where users struggle the most, with 32 per cent finding it difficult to identify trustworthy news content.
5
Decline in paying for local newspapers
While 40 per cent of Norwegians pay for digital news subscriptions, interest in local newspapers has dropped. Subscription bundles, such as Amedia's +Alt, which provides access to multiple local papers, are growing in popularity.
Norway faces massive potential job automation due to generative AI
This report by the Norwegian Board of Technology focuses on generative AI and its implications for Norwegian society.
Key findings
1
Generative AI’s rapid rise and societal impact
The breakthrough of large language models (LLMs), such as ChatGPT, has accelerated the development of generative AI. The report warns that while the technology promises to transform industries, it presents significant risks—including misinformation, data privacy concerns, and potential manipulation.
2
Significant workforce transformation
The potential to streamline and automate tasks may impact 155,000 jobs in the Norwegian public sector alone. However, the report highlights the need to balance job efficiency and upskill the workforce to adapt to new AI technologies.
3
Infrastructure and national AI models
Norway must invest in computational power and quality data infrastructure to fully harness AI's potential. The report calls for developing Norwegian-specific language models to maintain cultural and linguistic autonomy, ensuring digital tools align with national needs and languages.
4
Misinformation and disinformation risk
The report warns of an increasing threat of AI-generated deepfakes and misinformation, particularly in political contexts. Clear regulations on AI-generated content, especially during election campaigns, are recommended to preserve public trust and democratic integrity.
5
Policy recommendations
The expert panel proposes a coordinated effort across sectors to stimulate AI innovation, build infrastructure, and protect against AI misuse. Key measures include stronger national oversight, investments in AI technology, and public awareness to enhance digital resilience.
A small group drives most fake news on Twitter
Researchers at the Ben-Gurion University, Northeastern University, and Harvard University have studied the phenomenon of fake news "supersharers" on Twitter (now X) during the 2020 US presidential election.
Key findings
1
80 per cent of fake news shared by 0.3 per cent of voters
The study identified 2,107 individuals, referred to as supersharers, who accounted for 80 per cent of the fake news shared by 664,391 registered US voters. This small group reached 5.2 per cent of voters on Twitter, highlighting their outsized influence on the platform.
2
Supersharers are mostly older women and Republicans
Demographically, supersharers were found to be predominantly women, older adults, and registered Republicans. This contrasts with common stereotypes about social media manipulators being young or technologically savvy.
3
Fake news concentrated among supersharers’ networks
Supersharers played a critical role in spreading misinformation within their networks. Followers of supersharers were 2.5 times more likely to encounter fake news in their Twitter feeds than the average user.
4
Manual sharing dominates over bots
Despite concerns about automation, the study found that supersharers were not primarily using bots. Instead, they manually and persistently retweeted content, amplifying fake news through repeated actions.
5
Targeted interventions could limit fake news spread
The research suggests that platform interventions targeting supersharers or limiting retweets could significantly reduce exposure to fake news, as these individuals are central to its dissemination.
Loneliness in adolescence is linked to conspiracy beliefs in midlife
Researchers at the University of Oslo, Oslo Metropolitan University (OsloMet), and the University of Queensland have studied the long-term connection between loneliness and the development of conspiracy worldviews in adulthood. The research, conducted over nearly 30 years, focuses on a population sample from Norway.
Key findings
1
Loneliness in adolescence predicts conspiracist worldviews
Participants who reported higher loneliness during adolescence were more likely to hold conspiracy beliefs in midlife. This connection was evident even after controlling for variables like age, gender, education, and political orientation.
2
Loneliness increases over time for conspiracy believers
People whose loneliness increased significantly throughout their lives showed a higher likelihood of endorsing conspiracy theories. The study suggests that loneliness acts as a motivational factor driving individuals to seek meaning and social connection through conspiratorial thinking.
3
Ego defence mechanisms may fuel conspiracy thinking
The study theorises that lonely individuals use conspiracy beliefs to protect their self-esteem. By blaming external forces for their isolation, they can maintain a positive self-image and justify their loneliness.
4
Lack of social feedback reinforces conspiracy beliefs
Those who are lonely may lack the social support needed to challenge their developing conspiracy views. Without correction from trusted relationships, these individuals may turn to like-minded groups that reinforce their worldviews.
5
Loneliness interventions could help reduce conspiracist mindsets
The study suggests that addressing loneliness, especially during critical life stages, could be a way to mitigate conspiracy thinking and its societal impacts.
Football fuels 80 per cent of online hate in sports debates
Amnesty International Denmark have published a report analysing the prevalence and impact of hate speech and attacks in sports-related discussions on social media. The study focuses on Danish sports debates, especially on Facebook, based on over 5.5 million comments from sports media and club pages over a two-year period.
Key findings
1
Widespread hate speech in sports debates
About 182,000 comments (3.3 per cent of the total) across Facebook posts related to sports contain attacks. This includes everything from derogatory remarks to racist, sexist, and homophobic slurs, particularly targeting athletes and other users involved in the conversations.
2
Football debates are the most toxic
Football generates the highest number of attacks, with four per cent of comments on football-related posts being offensive. In contrast, other sports like swimming and badminton have significantly fewer hate comments. Football alone accounts for 80 per cent of all hate comments, underlining the intense emotions and polarised debates surrounding the sport.
3
Women and minorities face severe discrimination
The study highlights that women, LGBTQ+ individuals, and ethnic and religious minorities are frequent targets of hate speech. These attacks often involve stereotyping, discrimination, and exclusion, with women facing sexist comments and minorities being subject to both racism and xenophobia.
4
Hate extends to personal attacks
On personal Instagram profiles of professional footballers, the tone becomes even more vicious, with attacks often directed at players' personal lives, families, and appearance. The study also notes a worrying trend of sexualised and derogatory comments aimed at female partners of athletes.
5
Sports debates often turn political
Political and social issues regularly infiltrate sports discussions, with debates about war, minority rights, and controversial events further fuelling polarisation and hate. For example, political events like the World Cup in Qatar prompted a surge in hate comments, particularly involving discussions on human rights and LGBTQ+ issues.