Why Some People Trust Algorithms As Much As News Editors

The Hidden Reason People Trust Algorithm-Picked News

For many people, keeping up with the news doesn’t mean opening a newspaper or visiting a news website. It means scrolling through a social media feed and assuming the important stuff will show up on its own. Researchers call this the “news will find me” mindset, and a new study finds it may be quietly reshaping who, and what, people trust when deciding what’s real.

Published in the journal Social Media + Society, the study found that people who strongly hold this belief treat news recommended by social media algorithms, or even shared by online friends, with just as much trust as they place in professional journalists and editors.

Researchers didn’t stop at documenting that trust gap. They dug into why it exists, and landed on mental shortcuts: snap judgments people make about a source before reading a single word of the actual story. For people who already believe the news will come to them effortlessly, those shortcuts may carry unusual weight.

Source Cues Trigger Mental Shortcuts, Not Careful Reading

When a news story pops up in a social media feed, most people don’t have time or motivation to fully read and evaluate it. So the brain takes a shortcut: it looks at where the story came from and makes a fast judgment based on that alone.

Researchers have identified several of these shortcuts: an “authority shortcut,” where a story seems credible because a trained editor picked it; a “machine shortcut,” where a story seems objective because an algorithm chose it, on the logic that machines don’t carry personal biases; a “bandwagon shortcut,” where wide sharing signals something worth reading; and a “similarity shortcut,” where news feels more trustworthy coming from someone who seems like the reader.

None of these shortcuts are inherently bad. Given the volume of information flooding feeds daily, the brain simply cannot critically evaluate every headline. The problem arises when shortcuts replace thoughtful evaluation entirely, especially among people who weren’t putting much effort into seeking out news to begin with.

A 244-Person Experiment Put Three News Sources to the Test

To see how these shortcuts play out in real news consumption, researchers at multiple universities ran a controlled experiment with 244 U.S. participants recruited online, ranging in age from 19 to 77 with an average age of about 44. The group skewed slightly politically liberal.

Before the experiment began, participants answered questions measuring how strongly they held the “news will find me” belief. They were then randomly assigned to one of three conditions, each interacting with a mock news website called “Daily Times” showing a news story recommendation. The source label differed by group: one group was told the story came from professional news editors, another from their social media friends, and a third from an algorithm. Researchers tracked not just what participants said about the story, but what they actually did, including whether they clicked to read the full article first or liked, shared, or commented without reading it.

Algorithm Labels Made People Act Before Reading

Source labels shape behavior, the results showed, sometimes more than content itself.

Overall, participants rated editor-selected news as more credible, more important, and less biased than news attributed to friends, and saw editors as more of an authority than either friends or algorithms. An editor’s name attached to a story triggered the authority shortcut.

That pattern broke down for participants who scored high on the “news will find me” scale, though. Their opinion of algorithm- and friend-recommended news rose with how strongly they held the “news will find me” belief, while their opinion of editor-recommended news did not budge. A closer look at the data suggested why: these high scorers tended to view algorithms and friends as just as authoritative as editors, rather than reserving that trust for professional journalists alone.

A telling behavioral result also came out of the algorithm condition. When a story was labeled algorithm-recommended, participants were noticeably more likely to like, share, or comment on it without reading first, compared to stories from editors or friends. Nearly half in the algorithm condition acted before reading, versus roughly a third in the other two conditions.

The Findings Echo a Broader Pattern of Unread Sharing

This finding connects to a broader pattern in how news travels online, the researchers note. Prior research cited in the study found that a large share of news links shared on social media go unread entirely, and the current study suggests algorithmic recommendation cues may contribute to that behavior.

For people high in “news will find me” thinking, the risk compounds. Prior research cited by the authors shows these individuals are already more susceptible to accepting misinformation as credible, and when an algorithm serves them a story, trusting the machine to have picked something accurate cuts the motivation to double-check further. The researchers found that these individuals leaned harder on the machine shortcut with algorithm-recommended news, and that reliance was statistically linked to friendlier evaluations of the story and less interest in fact-checking it.

No single villain emerges here, not social media, not algorithms, not the people who rely on them. But it makes a pointed case that the assumption many people hold, that important news will simply find its way to them, may be steadily eroding the mental habits needed to tell real news from noise.

---

Disclaimer: This article summarizes findings from a peer-reviewed study and is intended for general informational purposes. It is not a substitute for reading the original research or for professional guidance on media literacy or misinformation.

---

---