The comment section is making a comeback. After a decade of newsrooms disabling comments to escape toxicity, a growing number are reopening them, betting that engaged readers are worth the moderation cost. The shift is driven by better AI moderation tools and a hard realization: comments drive loyalty, return visits, and the kind of community that social platforms can no longer deliver (NiemanLab, 2026).
For years, the calculus was simple. Comments were a moderation nightmare and a reputational risk. Newsrooms shuttered their sections one by one, pushing readers toward Facebook threads and Twitter replies instead. But as platforms like Meta pulled back from news distribution and algorithm changes buried publisher content, the industry watched its traffic pipeline collapse. The platforms that were supposed to replace comment sections became unreliable at best, hostile at worst.
Why newsrooms killed comments in the first place
The great comment shutdown of the 2010s was a response to genuine problems. Toxicity was rampant. Spam overwhelmed every thread. Moderation costs ballooned. A 2014 study found that even a small team of moderators could spend 20 to 30 hours a week reviewing comments on a single mid-sized publication. The math never worked. Most outlets chose to disable comments entirely rather than absorb that ongoing expense.
The UK saw this play out sharply. After the Online Safety Act passed in 2024, the regulatory burden on publishers managing user-generated content grew heavier. Newsrooms that had already shut down comments saw no reason to reopen them under tighter rules. But the regulation also created a framework for accountability that made AI moderation tools more acceptable to publishers and audiences alike (Reuters Institute DNR, 2025).
What changed to bring them back
Three forces converged to make comments viable again. First, AI moderation tools reached a level where they could filter abuse, spam, and misinformation at scale. Yahoo! News Japan launched an AI feature that detects offensive comments before they publish and asks users to reconsider. The result: offensive comments dropped by 24 percent after deployment (Reuters Institute DNR, 2025). Second, social platforms became unreliable traffic sources. Third, publishers realized that owning their audience relationships was no longer optional.
- AI moderation cuts the human cost of running comments by 70 to 80 percent
- Readers who comment visit 2 to 3 times more often than passive readers (NiemanLab, 2026)
- Newsrooms are building owned communities as social platforms deprioritize news content
- The UK Online Safety Act created regulatory clarity for managed comment sections
- Spain reports 57 percent of news organizations now use AI tools in their workflows (KPMG, 2025)
The engagement argument is hard to ignore
The business case for comments rests on one metric: time on page. Readers who leave a comment spend significantly longer on an article than those who simply read and leave. That extra engagement translates directly into ad revenue, subscription conversion, and the kind of audience loyalty metrics that advertisers value. For outlets struggling with declining traffic from social platforms, comments offer a way to capture and hold attention without paying for it.
Trust in news remains fragile worldwide. The Reuters Institute's 2025 Digital News Report found that only 30 percent of US respondents trust the news they consume, down 2 points from the prior year. In France, that number is just 29 percent. Comment sections can help rebuild trust by creating a transparent space where readers challenge claims, add context, and hold publications accountable. The conversation itself becomes a trust signal.
What good comment sections actually look like
The outlets doing this well are not just turning comments back on and hoping for the best. They are setting clear community guidelines, surfacing quality contributions through algorithmic ranking, and using AI to flag problematic content before it appears. The result is a comment section that adds context instead of noise. The Guardian, which never fully shut down its comments, serves as a model: moderated threads with clear rules produce substantive discussion that readers cite as a reason to visit.
The comment section is not a liability anymore. It is a retention tool.
— NiemanLab
Some outlets are going further. They are tying comment privileges to subscription status, creating tiered participation where paying readers get enhanced commenting features. Others are experimenting with upvote systems borrowed from Reddit and Hacker News, letting the community surface the best contributions. The thread ranking algorithm is becoming as important as the article itself.
AI moderation has real limitations
AI moderation is not perfect, and comment sections can still spiral into toxicity. The tools are good at catching explicit slurs, spam, and obvious abuse. They struggle with nuance, sarcasm, dogwhistles, and the kind of passive-aggressive toxicity that readers recognize but machines miss. The Japan model shows promise: the AI asks users to reconsider offensive comments, but it is ultimately up to the user whether to follow the advice. The human in the loop remains essential.
There is also the question of bias. AI moderation systems are trained on datasets that reflect existing power structures. They can over-flag comments from marginalized communities while missing coded language from dominant groups. Newsrooms that reopen their comment sections need to pair AI tools with human review, especially for sensitive topics like politics, race, and immigration where context matters most.
The German Press Council reported a record 86 serious violations of its Press Code in 2024, up from 73 the year before. Many complaints related to coverage of the Middle East conflict and terrorist attacks, areas where comment sections tend to be most volatile. Outlets that want healthy comment sections must invest in topic-specific moderation strategies rather than relying on a one-size-fits-all AI filter.
| Platform | AI Tool | Result |
|---|---|---|
| Yahoo! News Japan | Pre-posting detection | 24% drop in offensive comments (Reuters Institute, 2025) |
| The Guardian | Hybrid human-AI | Comments active with quality ranking |
| Meta (Facebook) | Community Notes model | Replaced fact-checking partnerships in US (Reuters Institute, 2025) |
| X (Twitter) | Community Notes | Crowdsourced moderation with mixed results |
The social media exodus forced the issue
The comment section comeback did not happen in a vacuum. It was forced by a series of platform decisions that cut publishers off from their audiences. Meta's decision to deprioritize news content on Facebook and Instagram reduced referral traffic to news sites by double digits across multiple markets. Google's search algorithm changes further eroded organic reach. Publishers who had relied on social platforms as their community layer suddenly found themselves building from scratch.
The numbers tell the story. In the US, 31 percent of respondents now share news via social media, messaging, or email (Reuters Institute DNR, 2025). That is a thin channel for publishers who once depended on platforms to distribute their content. Meanwhile, news avoidance is rising. In Croatia and Bulgaria, more than 60 percent of respondents reported avoiding news sometimes or often. Having a lively comment section gives readers a reason to engage with news rather than tune it out.
What this means for the future of online news
The comment section revival signals a broader shift in how publishers think about audience relationships. The era of outsourcing community to social platforms is over. Publishers who build their own comment sections, moderate them well, and create genuine spaces for reader participation will retain audience attention. Those who do not will continue bleeding engagement to platforms that no longer prioritize news.
The economics are straightforward. A comment section costs a fraction of what publishers pay to acquire readers through social media advertising. It costs even less than maintaining a presence on platforms that have shown they will change the rules at any time. For outlets chasing sustainable audience growth, the comment section is not a relic. It is a strategy.
This is not nostalgia for the early internet. Modern comment sections are heavily moderated, algorithmically ranked, and often tied to verified identities. They look nothing like the wild west of 2005. But they serve the same fundamental purpose: giving readers a voice and giving publishers a community they own. In a media world where trust is declining and attention is fragmented, that community is worth fighting for.
- Reuters Institute — Digital News Report 2025
- NiemanLab — Newsrooms are reopening their comment sections
- Reuters Institute — US market data, trust and engagement metrics
- Reuters Institute — Japan market data, Yahoo! News AI moderation
- The Dead Internet Theory Is Real. Bots Outnumber Us Now
- AI Slop Is Eating the Web Whole
- The 4-Day Workweek Went Corporate: Results Are In
- AI Companions Are Turning Real — and People Are Choosing Them Over Friends
- Digital Minimalism Is Booming. Here's Why People Are Logging Off
- Best AI Models 2026: The Definitive Ranking
Why are news sites reopening comments?
AI moderation cut costs by 70-80%, social platforms became unreliable, and publishers realized owning audience relationships was essential. Commenters visit 2-3x more often.
Does AI moderation work for comments?
Yahoo Japan reduced offensive comments by 24%. AI catches slurs but struggles with nuance. Best results pair AI with human review.
Do comments drive revenue?
Yes. Commenters spend longer on articles, boosting ad revenue and subscription conversion. Comments have become a retention tool.
Bottom line
This is not nostalgia for the early internet. Modern comment sections are heavily moderated, algorithmically ranked, and often tied to verified identities. They look nothing like the wild west of 2005. But they serve the same fundamental purpose: giving readers a voice and giving publishers a community they own. In a media world where trust is declining and attention is fragmented, that community is worth fighting for.
What we still don't know
This is a fast-moving story. We update the post as new facts land — and we'll flag it when we do.
Enjoyed this? Pay it forward
A sharp story is worth passing on. Share it with the people who read tech like it matters.