News outlets in poorer countries cover the world much as the leading outlets of rich countries do. Two explanations compete. One says they depend on Western news agencies for lack of resources. The other says they choose to copy prestigious outlets or whoever moves first. Observation cannot tell these apart. This study measured how closely 18 national outlets in Bangladesh, Kenya, South Africa, India, Malaysia and Nigeria followed four core outlets in how much attention they gave, which angle they took and when they published, on the wars in Ukraine and Gaza and two UN climate conferences in 2022-2024. It then ran a simulation in which language models acted as the outlets' editors and dependence, imitation and herding were switched on and off. Of nine tests, one held: outlets that relied more on wire copy published sooner after the core did. Tier and local stake showed no effect distinguishable from none. The herding test was supported and the imitation test was not. The simulation did not decide between the mechanisms, because the dependence rule, as fitted, reduced to publishing wire copy. Wire copy sets the clock of the periphery's foreign news. Whether it sets the agenda stays open.
Figures & tables
Tier
Issue
Cells (n, after the 20-story floor)
Coded articles (n)
Salience r (M, SD)
Framing similarity (M, SD)
Timing: first-story lag (Mdn days)
Timing: peak cross-correlation lag (Mdn weeks)
Periphery
Ukraine
9
433
0.33 (0.19)
− 0.12 (0.09)
− 0.12 (events 9,167)
0
Periphery
Gaza
9
449
0.19 (0.23)
− 0.10 (0.11)
− 0.04 (events 10,814)
0
Periphery
COP27/COP28
8
364
0.64 (0.29)
− 0.09 (0.09)
0.04 (events 131)
0
Periphery
All issues
26
1246
0.38 (0.29)
− 0.10 (0.09)
− 0.07 (events 20,112)
0
Semi-periphery
Ukraine
9
448
0.28 (0.27)
− 0.08 (0.06)
− 0.04 (events 11,530)
0
Semi-periphery
Gaza
9
450
0.11 (0.30)
− 0.11 (0.07)
− 0.03 (events 14,288)
0
Table 1. Following at Three Levels by Tier and Issue
Figure 1. Weekly Issue Shares, Core and Peripheral Outlets
Test
Predictor
Level
b
95% CI
p
Cells
Decision
H1
Tier (periphery)
Salience
0.103
[ − 0.148, 0.354]
1.000
52
not supported
H1
Tier (periphery)
Framing
− 0.010
[ − 0.065, 0.046]
1.000
52
not supported
H1
Tier (periphery)
Timing
− 0.026
[ − 0.132, 0.081]
1.000
37
not supported
H2
Wire reliance
Salience
− 0.020
[ − 0.267, 0.227]
1.000
52
not supported
H2
Wire reliance
Framing
0.019
[ − 0.066, 0.104]
1.000
52
not supported
H2
Wire reliance
Timing
0.175
[0.084, 0.267]
.002
37
supported
Table 2. Mixed Models of Following on Tier (H1), Wire Reliance (H2) and Tier × Local Stake (H3), With the Holm Decisions and the Two Falsifiers
Figure 2. Following by Tier and Local Stake
Configuration
Error: salience
Error: framing
Error: timing
Mean error
vs baseline (M, SD)
vs best other single rule (M, SD)
Signs reproduced
Bootstrap share
Within-outlet error
Baseline (no rule)
1.13
1.06
0.79
1.00
—
—
0.50
—
0.93
Dependence
1.25
0.95
0.28
0.83
− 0.17 (0.01)
− 0.08 (0.01)
0.71
0.45
0.93
Imitation
0.85
1.08
0.79
0.91
− 0.09 (0.00)
0.08 (0.01)
0.50
—
0.88
Herding
1.22
1.09
1.69
1.33
0.34 (0.02)
0.51 (0.02)
0.53
—
1.39
Dependence + imitation
1.19
1.01
0.43
0.88
− 0.12 (0.01)
—
0.44
—
0.88
Dependence + herding
1.30
0.98
7.29
3.19
2.20 (0.45)
—
0.62
—
1.42
Table 3. Out-of-Sample Fit of the Eight Simulation Configurations
Figure 3. Standardized Out-of-Sample Error by Level and Configuration
News outlets covering the same event differ in what they emphasize and omit. Same Story, Different Lens is an experimental research prototype for comparing these choices. It collects world and politics coverage from 17 outlets, groups reports about the same event, and produces structured analyses of sentiment, political lean, framing, loaded language, possible omissions, and model-estimated proportions of factual claims and opinion. Loaded-language entries include quotations and neutral rewordings. Three model roles (agents) handle article analysis, assessment of source support, and cross-outlet briefing generation. Analyses are stored in a JSONL archive. The evaluation compares outlet-level lean estimates with approximate, predefined reference ratings and summarizes the Critic's model-assigned grounding scores. These measures provide preliminary diagnostics. A grounding score reflects the model's assessment of support in the supplied text and does not establish factual accuracy or constitute independent verification.
Affan Yasin, Yueqi Lyu, Rubia Fatima +2
Academy of AI and Advanced Technology (AIAT), Xi’an Jiaotong-Liverpool University (XJTLU), Suzhou, China · Department of Computer Science and Technologies, Emerson University, Multan, Pakistan.
This paper examines the diffusion of media slant. We document the influence of Fox News Channel (FNC) on the partisan slant of local newspapers in the U.S. over the years 1995-2008. We measure the political slant of local newspapers by scaling the news article texts to Republicans' and Democrats' speeches in Congress. Using channel positioning as an instrument for viewership, we find that higher FNC viewership causes local newspapers to adopt more right-wing slant. The effect emerges gradually, only several years after FNC's introduction, mirroring the channel's growing influence on voting behaviour. A main driver of the shift in newspaper slant appears to be a change in local political preferences.
State media outlets often project their governments' narratives to target both domestic and foreign audiences. Because this content is plentiful and free to access, it appears widely in the web corpora that large language models (LLMs) are trained on and that retrieval systems draw from. However, the prevalence of state media in these corpora and its potential influence on LLM outputs remain largely unexplored. In this paper, we introduce WorldStateMedia, an open dataset of 118 million state media articles from 136 countries in 114 languages. We extract these articles from existing pretraining corpora and annotate each outlet with level of state control from State Media Monitor. Using WorldStateMedia to audit these corpora, we find that, in English web crawls, the share of state media pages from editorially controlled (vs. independent) outlets rose from 39% in 2013 to 93% in 2025. In several low-resource languages, state media make up a substantial share of all web data, such as 17.8% for Uyghur and 7.8% for Belarusian. To characterize the differing narratives of news outlets, we compare Russian and Ukrainian state media in English and in their native languages, revealing asymmetric topical focus and opposing framings of contested issues such as Crimea. Furthermore, in a retrieval-augmented generation (RAG) experiment on territorial disputes, retrieving a country's state media shifts model answers toward that country's claims, most strongly for English-language (foreign-targeted) media. We release WorldStateMedia to support multilingual, cross-national insights on state media at https://huggingface.co/collections/manestay/worldstatemedia
Bryan Li, Samar Haider, Grace Kim +6
Google · Northwestern University · University of Pennsylvania +2