62% of people now worry about telling real news from fake online, while only 37% trust the news they see, the lowest figure recorded since 2015. That 25-point gap between concern and confidence is the central tension in media data right now. Concern rose 4 points in a single year, and trust in news delivered through AI chatbots is lower still at 20% even as weekly chatbot news use climbs from 7% to 10%. The figures below come from the Reuters Institute Digital News Report, covering 48 markets with roughly 2,000 respondents each.
TL;DR
- 62% are concerned about distinguishing real from fake news (Reuters Institute)
- That is up 4 percentage points year on year (Reuters Institute)
- Trust in news overall is 37%, the lowest since 2015 (Reuters Institute)
- US trust in news is 25% (Reuters Institute)
- Among right-leaning Americans it is 15% (Reuters Institute)
- Trust in news from AI chatbots is 20% (Reuters Institute)
- Weekly AI chatbot use for news rose from 7% to 10% (Reuters Institute)
- Among under-35s it reaches 16% (Reuters Institute)
- Only 1% say AI is their main news source (Reuters Institute)
- Social media and video networks reach 54% for news (Reuters Institute)
- Facebook is the largest single news platform at 43% (Reuters Institute)
- 42% report avoiding news, unchanged year on year (Reuters Institute)
- High interest in news fell to 46% from 59% in 2021 (Reuters Institute)
- 17% pay for online news across a 20-country basket (Reuters Institute)
1. Concern About Fake News Is Rising
The concern figure moved meaningfully in one year, which is unusual for an attitudinal measure. 62% of respondents say they are concerned about distinguishing real from fake news online, up 4 percentage points, with concern highest in Nigeria and Kenya, followed by the United Kingdom, Australia, and Portugal.
The country ranking cuts against the assumption that this is primarily a rich-democracy anxiety. Nigeria and Kenya lead, and both have recent experience of coordinated online manipulation around elections alongside high mobile-first social media use, which suggests concern tracks exposure rather than media literacy in the abstract. The presence of the UK, Australia, and Portugal immediately behind them shows the worry is not confined to any one media environment. A 4-point annual rise in a measure already above 60% indicates the problem is being perceived as worsening, whatever the underlying reality of misinformation volume.
| Metric | Value | Source |
|---|---|---|
| Concerned about fake news | 62% | Reuters Institute |
| Year-on-year change | up 4 points | Reuters Institute |
| Implied prior year figure | approx. 58% | Derived from Reuters Institute figures |
| Highest concern | Nigeria and Kenya | Reuters Institute |
| Next highest | UK, Australia, Portugal | Reuters Institute |
| Markets covered | 48 | Reuters Institute |
| Respondents per market | approx. 2,000 | Reuters Institute |
| Approximate total sample | 96,000 | Derived from Reuters Institute figures |
Source: Reuters Institute Digital News Report 2026 executive summary.
2. Trust Is at a Record Low
The confidence side of the equation is worse than the concern side. Trust in news overall stands at 37%, the lowest level recorded since 2015, with the United States at 25% and right-leaning Americans at 15%.
The internal American spread is the most striking figure in the dataset. A 10-point gap between the national average and one political segment means trust is not simply low but unevenly distributed along partisan lines, which changes what a remedy would look like. A general credibility problem could in principle be addressed by better journalism; a partisan one implies that a substantial group has largely exited the shared information environment and will not be reached by improvements to outlets they have already rejected. At 15%, roughly six in seven right-leaning Americans do not trust the news they encounter.
| Metric | Value | Source |
|---|---|---|
| Trust in news overall | 37% | Reuters Institute |
| Historical position | lowest since 2015 | Reuters Institute |
| US trust in news | 25% | Reuters Institute |
| Right-leaning Americans | 15% | Reuters Institute |
| Gap, US average to right-leaning | 10 points | Derived from Reuters Institute figures |
| Gap, global to US | 12 points | Derived from Reuters Institute figures |
| Trust in AI chatbots for news | 20% | Reuters Institute |
| Gap between concern and trust | 25 points | Derived from Reuters Institute figures |
Moderation context sits in our content moderation statistics. Source: GIJN on the 2026 Digital News Report.
3. Where People Actually Get News
The distribution channel data shows three roughly equal pillars rather than the collapse of traditional media often described. Social media and video networks reach 54% for news, television 52%, and news organisations’ own websites and apps 51%.
Those figures are close enough that the ordering matters less than the fact of parity. What has changed is not that one channel replaced another but that no channel commands a majority position, and audiences now distribute across all three. Within social platforms the concentration is clearer: Facebook remains the single largest at 43%, well ahead of YouTube at 34%, Instagram at 26%, and TikTok at 20%. Facebook’s continued lead is worth noting because it contradicts the widespread assumption that news consumption has migrated to newer video-first platforms; it has diversified across them without abandoning the incumbent.
| Metric | Value | Source |
|---|---|---|
| Social media and video networks | 54% | Reuters Institute |
| Including AI chatbots | 56% | Reuters Institute |
| Television news | 52% | Reuters Institute |
| Owned news websites and apps | 51% | Reuters Institute |
| Facebook for news | 43% | Reuters Institute |
| YouTube for news | 34% | Reuters Institute |
| Instagram for news | 26% | Reuters Institute |
| TikTok for news | 20% | Reuters Institute |
| Weekly online video consumption | 77% | Reuters Institute |
Platform-level context sits in our social media statistics. Source: NewscastStudio on the 2026 report release.
4. AI Chatbots Enter the News Diet
The AI numbers describe rapid relative growth from a small base, which is easy to report in either direction. Weekly use of AI chatbots for news rose from 7% to 10%, reaching 16% among under-35s, while only 1% say AI is their main source of news.
Both framings are true and the second one matters more for now. A jump from 7% to 10% is roughly 43% growth in a year, and the under-35 figure at 16% shows where it is concentrated, which together make a reasonable case that this becomes a mainstream channel. But 1% naming it as a primary source means chatbots are currently additive rather than substitutive. The trust figure complicates the growth story further: at 20%, confidence in AI-delivered news sits 17 points below already-depressed trust in news generally, meaning adoption is running ahead of belief. People are using a channel they do not trust, which is not a stable configuration.
| Metric | Value | Source |
|---|---|---|
| Weekly AI chatbot use for news | 10% | Reuters Institute |
| Prior year | 7% | Reuters Institute |
| Relative growth | approx. 43% | Derived from Reuters Institute figures |
| Under-35 use | 16% | Reuters Institute |
| AI as main news source | 1% | Reuters Institute |
| Trust in news from AI chatbots | 20% | Reuters Institute |
| Trust in news overall | 37% | Reuters Institute |
| Trust gap for AI | 17 points below general news | Derived from Reuters Institute figures |
Accuracy context sits in our AI hallucination statistics. Source: Reuters Institute on AI and the future of news.
5. Creators Take a Large Share
The creator economy has moved from the margins of news consumption to a substantial share of it. 46% get news from creators of some type and 27% from news-focused creators specifically, while only 3% rely on creators alone.
The gap between 46% and 3% is the important structure here. Creators have achieved wide reach without displacing anything, functioning as an additional layer on top of existing habits rather than a replacement for them. That said, 27% getting news from news-focused creators puts individual personalities on a comparable footing with major platforms, and unlike institutional outlets they carry no editorial infrastructure, correction policy, or accountability mechanism. The accountability question is not whether creators are less accurate than newsrooms, which the data does not address, but that a quarter of the audience now receives news through a channel with no structural mechanism for being wrong in public.
| Metric | Value | Source |
|---|---|---|
| Get news from creators, any type | 46% | Reuters Institute |
| From news-focused creators | 27% | Reuters Institute |
| Rely solely on creators | 3% | Reuters Institute |
| Ratio, any use to sole reliance | approx. 15 to 1 | Derived from Reuters Institute figures |
| Comparison, news creators to YouTube | 27% versus 34% | Derived from Reuters Institute figures |
| Function in the news diet | additive | Derived |
| Editorial infrastructure | generally absent | Derived |
| Markets covered | 48 | Reuters Institute |
Source: Edelman on the 2026 Digital News Report.
6. Disengagement Outpaces Avoidance
The engagement data points at something slower and more consequential than active rejection. 42% report avoiding news, unchanged year on year, while high interest in news fell to 46% from 59% in 2021 and casual or passive users rose from 16% to 25%.
Avoidance being flat while interest drops 13 points over five years means the shift is not people angrily turning away but drifting away. Active avoidance requires caring enough to make a decision; the growth is in the casual and passive category, up nine points, which describes people who simply stopped seeking news out. That distinction has practical consequences, because avoidance responds to changes in tone and volume of coverage while passive drift responds to nothing in particular. The payment data reinforces the picture: 17% pay for online news across a 20-country basket of wealthier markets, with Norway at 40% and Sweden at 32% as outliers rather than leading indicators.
| Metric | Value | Source |
|---|---|---|
| News avoidance | 42% | Reuters Institute |
| Change year on year | unchanged | Reuters Institute |
| Highly interested in news | 46% | Reuters Institute |
| Same figure in 2021 | 59% | Reuters Institute |
| Decline since 2021 | 13 points | Derived from Reuters Institute figures |
| Casual or passive users | 25% | Reuters Institute |
| Prior figure | 16% | Reuters Institute |
| Paying for online news, 20-country basket | 17% | Reuters Institute |
| Norway | 40% | Reuters Institute |
| Sweden | 32% | Reuters Institute |
Source: European Federation of Journalists on the 2026 report.
Summary: Misinformation and Media Trust by the Numbers
| Metric | Value | Source |
|---|---|---|
| Concerned about fake news | 62% | Reuters Institute |
| Year-on-year change in concern | up 4 points | Reuters Institute |
| Trust in news overall | 37% | Reuters Institute |
| Historical position of that figure | lowest since 2015 | Reuters Institute |
| US trust in news | 25% | Reuters Institute |
| Right-leaning Americans | 15% | Reuters Institute |
| Trust in AI chatbots for news | 20% | Reuters Institute |
| Weekly AI chatbot news use | 10% | Reuters Institute |
| Prior year | 7% | Reuters Institute |
| Under-35 AI use | 16% | Reuters Institute |
| AI as main news source | 1% | Reuters Institute |
| Social media and video networks | 54% | Reuters Institute |
| Television news | 52% | Reuters Institute |
| Owned websites and apps | 51% | Reuters Institute |
| Facebook for news | 43% | Reuters Institute |
| YouTube for news | 34% | Reuters Institute |
| News from creators, any type | 46% | Reuters Institute |
| News avoidance | 42% | Reuters Institute |
| Highly interested in news | 46% | Reuters Institute |
| Paying for online news | 17% | Reuters Institute |
Methodology and Sources
- All figures come from the Reuters Institute Digital News Report 2026, covering 48 markets with approximately 2,000 respondents per market (Reuters Institute executive summary, Reuters Institute on AI and news).
- Independent coverage and analysis of the same release were cross-checked against the primary publication (GIJN, NewscastStudio, Edelman, European Federation of Journalists).
- Data watch: this is an online survey, which systematically under-represents people with limited internet access and makes figures from lower-income markets less comparable to those from high-connectivity ones. All measures are self-reported attitudes and recalled behaviour rather than observed usage, so platform shares reflect what people believe they do. The concern measure asks about worry regarding fake news, not exposure to it, and rising concern is therefore not evidence that misinformation volume increased. The payment figure covers a 20-country basket of wealthier markets, not all 48, and is not comparable to the global measures alongside it. Political segmentation is available only for some markets, so the US partisan breakdown has no global equivalent here. Rows marked as derived are arithmetic on published figures.
- Last updated: August 2, 2026. We update this roundup quarterly, and the next major refresh is expected when the Reuters Institute publishes its next annual Digital News Report.