How do the sentiments expressed in press article headlines about the future of artificial intelligence vary according to thematic content, geographic regions, and temporal evolution, and what lessons can be drawn from this for strategic marketing?
Files
Gelhay_35201800_2025.pdf
Closed access - Adobe PDF
- 2.05 MB
Details
- Supervisors
- Faculty
- Degree label
- Abstract
- This thesis analyzes how international media portray artificial intelligence (AI) through newspaper headlines from 2010 to 2025. Using sentiment analysis tools—Orange Data Mining (with VADER) for text preprocessing and scoring, and SPSS for statistical validation—this study examines emotional tone variations based on topic, time, and geography. The corpus includes headlines from The Guardian, The New York Times, and English-language Asian press. Four hypotheses were tested: (1) ethical headlines are more negative than economic ones; (2) AI coverage increases over time while sentiment becomes more negative; (3) media discourse is polarized between utopian and dystopian narratives; (4) Global North media are more critical than Global South outlets, which are more optimistic. Findings show partial support: The Guardian confirms hypothesis (1), but not the NYT; hypothesis (2) is strongly confirmed with rising AI visibility and declining sentiment; hypothesis (3) is not supported—neutral tones dominate; hypothesis (4) is validated, with Asian press showing a markedly more positive tone. This study highlights key insights for corporate communication strategies and advocates for geographically adapted messaging. It also suggests further research incorporating qualitative methods and more diverse linguistic sources.