Could adding AI to online news comments make everyone's conversation feel worse, even for people who never use it?
Can AI involvement in news discussion reduce perceived quality without reducing use?
This explores whether bringing AI into online discussion of the news can make conversations feel worse to people while they keep participating as much as before, or even more.
This explores whether AI in online discussion can make conversations feel worse while people keep taking part. The short answer from the corpus is yes. In fact, it's stronger than the question suggests: use doesn't just hold steady, it goes up. In a 680-person experiment, AI commenting tools produced longer comments and more participation, yet readers judged the discussion as more generic and less authentic Do AI writing tools improve online discussion or degrade it?. The detail most worth knowing is that the drop in quality spread to everyone. Even conversations between people who never touched the AI felt worse. Once some voices in a thread are machine-assisted, the whole thread seems to lose credibility.
Why would quality fall while activity rises? One explanation is that AI changes how writers come across. A study of nearly 3,000 writers found AI help shifted every one of 29 measured traits. Writers came across as more confident, more extreme, more agreeable and more privileged Does AI writing assistance change how readers perceive the writer?. Writers rarely push back: they edited AI text only 23% of the time, and their edits left about 96% of it unchanged Do writers actually edit AI-generated text before publishing?. So the extra volume is real, but it arrives in a voice that isn't quite the writer's own, and readers can sense it.
A second explanation looks at what a discussion actually is. Some notes argue that AI's threat to social platforms isn't false content or the wrong tone. It's the loss of a conversational style, the sense of one person actually addressing another Does AI threaten social media's conversational function?. AI text carries the surface signs of an utterance without the event behind it, so readers do the interpretive work to treat it as a real exchange Does AI generate genuine utterances or just text patterns?. That's why engagement metrics can keep climbing: people are still replying and reacting. But what they're replying to is thinner. Moderation, fact-checking and ranking algorithms can't fix this, because nothing in the content is factually wrong.
The longer-term risk is to reputation. AI posts can win attention by being comprehensive, but that attention doesn't build any lasting speaker's standing Does AI content displace human influencers on social media?. A news discussion space could stay busy and monetizable while losing its role as the place where trusted human voices emerge. Scale up the volume and you get what one note calls epistemic hyperinflation: more claims are produced than anyone can evaluate, so each one counts for less Can AI generate knowledge faster than humans can evaluate it?.
One useful contrast: when people use AI chatbots to get news directly, trust and use move together. Chatbot users are far more likely to trust chatbot news than non-users are, which suggests trust gates whether people adopt them at all Does trust in AI chatbots drive news-seeking behavior?. So the split between quality and use may be specific to discussion, where AI is mixed into a shared space people are already in, rather than a tool they choose. A caveat: the corpus has one direct experiment on this. The mechanisms above are well supported, but nothing here yet tracks whether use eventually falls once the perceived drop in quality builds up over time.
Sources 8 notes
In a 680-participant experiment, AI-assisted commenting tools produced longer comments and higher participation rates, yet readers perceived the content as generic and less authentic. The perceived decline in quality extended even to conversations among users who did not use the AI tools.
A study of 2,939 writers and 11,091 readers found AI assistance shifted every tested dimension—29 total—toward extremism, confidence, quality, agreeableness, and perceived privilege. Distortions were statistically significant and directional, not random noise.
Writers edited AI-generated paragraphs only 23% of the time, with edits averaging 96% similarity to the original. This means AI's opinionated and distorted voice propagates with minimal human filtering before publication.
AI-generated posts drain social media's function as a conversational medium because they lack the structure of genuine address and mutual orientation. This threat operates below the level where content moderation, fact-checking, and recommender adjustment can reach.
AI output carries communicative markers inherited from training data but lacks the event structure that produces actual utterances. Users supply the missing orientation through interpretive labor, creating a pseudo-event with structure only on the human side.
Show all 8 sources
AI-generated posts capture engagement through comprehensiveness but accrue social proof without building any speaker's sustained reputation. This displacement compounds over time, eroding the platform's core function of promoting legitimate human voices while monetization continues.
AI produces knowledge faster than human judgment can verify it, collapsing epistemic confidence just as monetary hyperinflation collapses purchasing power. The gap self-reinforces because evaluation tools are themselves AI-generated, trapping the system in acceleration.
A 45-market survey finds AI chatbot use for news rose to 10% globally, with trust in chatbots correlating more strongly with use than trust in social media does. Among chatbot users, 44% trust news from them versus 17% of non-users, suggesting trust gates deliberate adoption.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- Understanding Reader Perception Shifts upon Disclosure of AI Authorship
- Penalizing Transparency? How AI Disclosure and Author Demographics Shape Human and AI Judgments About Writing
- Blissful (A)Ignorance: People form overly positive impressions of others based on their written messages, despite wide-scale adoption of Generative AI
- Measuring and Mitigating Persona Distortions from AI Writing Assistance
- The Impact of Generative AI on Social Media: An Experimental Study
- "It was 80% me, 20% AI": Seeking Authenticity in Co-Writing with Large Language Models
- What Influences Readers' and Writers' Perceived Necessity of AI Disclosure?
- AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuances