Young adults use AI the most but trust it the least — could their news and social feeds be why?

This explores why young adults use AI more yet feel worse about it, and whether what they read and watch explains the gap. The direct answer is that the collection has no study of young adults' attitudes toward AI and none that links media diets to AI optimism or pessimism, so it can't settle the question. What it does offer are two competing explanations worth keeping apart, plus a caution about how to read group differences at all.

The media-diet explanation has a real mechanism behind it. Recommendation feeds aren't neutral pipes. They work as persuasion infrastructure: what the algorithm rewards changes what creators produce, how people are connected pushes their opinions toward each other, and these effects build on themselves (How do recommendation feeds shape what people see and believe?). If young adults get most of their news through feed-driven platforms, an anxious or critical story about AI could spread and harden within their networks quite separately from how they actually experience the tools. That makes media diet a plausible cause. But the research here describes how feeds shape belief in general, not beliefs about AI.

The second explanation is the one the question may be overlooking: heavy use itself might produce the pessimism. Several notes suggest that sustained, close contact with AI reveals its costs. A four-month EEG study found that brain connectivity and memory of one's own work dropped as people relied more on LLMs (Does AI assistance weaken our brain's ability to think independently?). A year-long study of Character.AI users linked sustained companion use to lower well-being, mostly because it replaced face-to-face time (Does sustained engagement with AI companions harm well-being?). And AI's persuasive edge fades over repeated conversations, while a human persuader's holds steady (Does AI persuasiveness fade across repeated conversations with the same person?). In other words, familiarity can wear down AI's appeal. On this reading, heavier users may be more pessimistic because of what they've noticed, not what they've scrolled past. That would make the 'despite' in the question a 'because.'

A third idea is less obvious and concerns method. When Indian writers accepted more AI suggestions than American writers, the researchers argued that the difference reflects cultural patterns of trust and shouldn't be statistically removed as noise (Is higher AI use by Indian writers a confound to control?). The same logic applies to age. Young adults' mix of heavy use and pessimism may be one coherent generational stance (dependent on the tools and wary of them at the same moment) rather than a puzzle that a hidden variable like media diet needs to explain away.

The question that would actually separate these explanations is whether pessimism among young adults tracks how much they use AI, or how much they consume feed-driven media about it. The collection points to both mechanisms but doesn't test either one against attitudes. That gap is a good candidate for adding sources.