The latest episode of Reality checks with Brian Reitz features Vince Nero, Director of Content Marketing at BuzzStream, covering AI news summary consumption habits of Americans, delving into trust, adoption, and behaviors.
AI-driven news summaries are increasingly prominent in the search engine results and across media publishing sites. But does the U.S. population actually trust them?
Vince Nero, Director of Content Marketing at BuzzStream collaborated with YouGov on a five-question survey of 1,000 Americans to find out.
“BuzzStream is very much in the PR space as a tool, so I am constantly trying to understand how people interact with this stuff for the news,” said Nero. “I think it has big implications for how people get their information, beyond just PR.”
This survey repeats a BuzzStream and YouGov survey from January 2026, so Nero was asked to predict both the current state of AI news consumption as well as shifts from the previous collaboration.
The article below provides further analysis to support the talk show interview, answering:
- Do Americans trust AI news summaries more than traditional news?
- Are there demographic differences in trust of AI news summaries?
- How has adoption of AI news summaries changed since January 2026?
- How has behavior with AI news summaries change among readers?
- Which social networks have higher concentrations of users adopting AI news tools?
U.S. consumers trust traditional news more than AI news summaries
Americans remain less trustworthy of AI-generated news summaries compared with traditional news sources.
Three in five U.S. adults (60%) say they trust AI summaries less than traditional news broadcasts or newspapers, including 42% who trust them much less.
This lack of trust in AI summaries (or increased trust in traditional news), reflects a slight shift from January sentiment, where 57% said they trusted traditional news more.
“We’re going to start to see dipping in trust and usage,” said Nero. “That’s at least my prediction across the board.”
Demographic differences in trust in AI news summaries vs. traditional media sources
Digging deeper into demographics for the September results, traditional news sources are more trusted than AI-generated news summaries across every demographic group measured, but the difference varies.
Gen Z adults are the most likely generation to trust traditional news (65%), followed by Baby Boomers (63%), compared with 54% of Millennials and 55% of Gen X.
Differences appear across race and ethnicity: 62% of White Americans trust traditional news more, compared with 54% of Black Americans and 48% of Hispanic Americans.
Hispanic adults are instead more likely than Americans overall to trust AI and traditional news equally (21% vs. 13% U.S. gen pop), while Black adults are the segment most likely to say they trust AI summaries more than traditional news (8% vs. 5% overall).
There is little difference between men and women, with around six in ten in each group saying they trust traditional news more.
While differences in trust exist, the overarching sense among Americans is that AI news summaries are not as trustworthy as traditional news, which has notably encountered its own challenges with trust.
Most Americans say they never or rarely use AI summaries for news
Most Americans still do not rely on AI tools as a regular source of news.
In January 2026, 61% of U.S. adults said they never or rarely get news from tools such as Google AI Overviews, social media AI, or AI chatbots. That number has increased to 64% in September.
Around three in ten (29%) say they sometimes, often, or always consume news this way.
The Americans who “always” use AI summaries to consume news doubled from 1% in January 2026 to 2% in September 2026.
AI news summary readers report catching hallucinations slightly less, and click through to sources slightly more than at the start of the year
Among the Americans who said they use AI-generated summaries for news, reports of encountering "hallucinated" information have declined since the start of the year, down from 72% noticing sometimes or more, to 64% just eight months later.
While this is a positive trend for AI companies, the data shows that the majority of users still trust traditional news outlets more.
Despite this, most AI summary readers do not read the original source material.
37% say they always or often click through to the source material, with the remaining AI news consumers doing so less frequently or never. This tendency has remained consistent with usage behaviors from the initial survey.
WhatsApp, X, and LinkedIn users lead across AI news adoption
Around three in ten U.S. adults (28%) say they have used an AI chatbot to explain a complex news topic, either regularly or on specific occasions.
This adoption is higher among users of WhatsApp (40%), X (39%), and LinkedIn (36%), while around a third of Snapchat (35%) and Reddit (34%) users have also turned to AI chatbots news summaries.
Regular use remains relatively uncommon, peaking at 10% among X users.
“I wonder with X, if people are looking at this question and saying, ‘I use Grok on X... I’m using the AI platform or the AI chatbot on the platform,” said Nero.
Across all social networks with qualifying sample size, adoption of AI among users meets or exceeds the U.S general population, with the exception of one: Pinterest.
Does this mean Pinterest users are less interested in AI, or less interested in news? That may require another survey!
