Desember 2023
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
The Paradox of AI Disclosure in News Trust
Researchers from the Hubbard School of Journalism & Mass Communication at the University of Minnesota and the Oxford Internet Institute at the University of Oxford have explored the impact of disclosing AI involvement in news production on audience trust.
Key Findings
1
AI-generated news perceived as less trustworthy
Despite no significant difference in perceived accuracy or fairness, news labelled as AI-generated was deemed less trustworthy by audiences.
2
Trust and knowledge about journalism affect perceptions
The negative impact on trustworthiness was more pronounced among participants with higher pre-existing trust in news and those with greater knowledge about journalism.
3
Source disclosure mitigates trust issues
Disclosing the sources used to generate AI-produced content significantly counteracted the negative perceptions of trustworthiness.
4
Complex attitudes towards AI in journalism
While AI-generated content is viewed sceptically, there is a strong preference for transparency about AI use in news production.
Evolving Dynamics of Norway's Media Economy
In this report, The Norwegian Media Authority offers a comprehensive analysis of the Norwegian media industry's economic performance over five years (2018-2022), revealing shifts in revenue streams, profitability, and sectoral growth patterns.
Key Findings
1
Slight revenue increase with a digital shift
The media industry saw a modest revenue increase, with total earnings rising by 1.8 per cent from 2021 to 2022. This period marked a pivotal shift towards digital, as user-generated revenues outpaced advertising revenues.
2
Challenges in profitability
Despite revenue growth, the industry faced profitability challenges, with operational costs increasing by 13 billion NOK (+5%) in 2022. These increases were primarily attributed to investments in broadcasting rights, market positions, and digital development, with the newspaper segment alone experiencing a 762 million NOK (+68%) cost increase.
3
Digital development as a key focus
The ongoing transition to digital platforms is evident, with significant investments in digital development and strategic acquisitions to enhance digital revenue streams. This strategy indicates media companies' efforts to diversify and strengthen their positions in a competitive market.
85 per cent concerned about the impact of disinformation
UNESCO has conducted a survey encompassing inhabitants in 16 countries, offering a global perspective on the challenges the digital information landscape poses, particularly in the context of upcoming elections.
Key Findings
1
Social media as a primary information source
A significant shift has been observed, with social media surpassing traditional media as the primary news source globally. However, trust in social media for credible information remains lower than in conventional media outlets.
2
Widespread concern over disinformation
A staggering 85 per cent of respondents expressed concern about the impact of disinformation on their societies.
3
Hate speech prevalence
Two-thirds of internet users frequently encounter hate speech online, with LGBT+ individuals and ethnic/racial minorities identified as the most common targets.
4
Call for regulatory action
There is broad consensus (89%) among the surveyed population for governments and regulators to mandate social media platforms to implement trust and safety measures during elections to safeguard electoral integrity.
AI has Limited Influence on Correcting Human Bias
Researchers from Ohio State University, Carnegie Mellon University, Columbia University, and Shell have conducted a pilot study to investigate if AI can help reduce human perceptual biases in visual tasks.
Key Findings
1
Faster decision-making with AI assistance
Participants were quicker in completing tasks with AI aid than without, indicating that AI can improve efficiency in visual tasks.
2
Persistence of human bias
Despite faster decision-making times, human biases, particularly the "pull-down" effect, remained largely unaffected by AI assistance, suggesting a limited influence of AI on correcting innate human biases in this context.
3
Impact of AI response time
The study found no significant effect of AI response time delays on the accuracy of human decisions, challenging assumptions about the role of timely AI input in improving decision accuracy.
Massive Use of News Content for AI Training
This white paper, published by the News/Media Alliance, looks into the complex issue of copyright infringement by generative artificial intelligence (GAI) systems through the unauthorised use of news, magazine, and digital media content for training purposes.
Key Findings
1
Massive copying of creative content
GAI developers extensively copy and utilise online news, magazine, and digital media content to train their models, often without authorisation, which infringes on the copyright owner's exclusive rights.
2
Evidence of overrepresentation
Technical analyses reveal that news and media content is disproportionately used in training sets. GAI models can memorise and reproduce this content, challenging the notion of non-infringing fair use.
3
Harming publishers
This unauthorised use severely harms publishers by diminishing web traffic, undercutting ad revenue, and weakening the trust and brand relationships built with readers over decades.
4
Legal and ethical challenges
The paper argues that the fair use doctrine does not justify the extensive copying of expressive works for GAI training, emphasising the need for a legal framework that ensures fair compensation for content creators.
Emerging Communities of Immigration Scepticism on Social Media
Researchers from Oslo Metropolitan University and the Norwegian Centre for Social Research explore how individuals sceptical of immigration in Norway create informal social media networks. These networks offer support and a space for sharing views, often in contrast to mainstream societal opinions.
Key Findings
1
Creation of alternative online communities
Sceptics of immigration policy form online groups to share information and support, especially when facing criticism in broader society.
2
Strategic participation
Participants carefully manage their online presence to engage with like-minded individuals while avoiding negative repercussions in their professional and personal lives.
3
Facebook's central role
Facebook is essential for these individuals to connect, share content, and find community, balancing active discussions and passive observation.
4
Experiencing stigmatisation
Individuals feel stigmatised for their views in daily life, prompting them to seek out online spaces where their opinions are validated.
5
Navigating extremist associations
Active community members work to present their views moderately, avoiding association with extremist rhetoric through cautious moderation and self-presentation.