You scroll through a thread on Hacker News—a community that once felt like a secret club of brilliant minds. But now, every other comment is [dead]. Grayed out. Gutted. Conversations that look like a battlefield after the war is over.
You feel it, don’t you? That slow, sinking frustration. The same nostalgia that makes you miss the old internet, when you could trust a stranger’s opinion without wondering if it was written by a bot. We’re watching the slow, public death of open, anonymous forums—and the enemy isn’t malice. It’s volume.
In the last week alone, Hacker News has been hammered by AI bots. Not clever ones. Not persuasive ones. Just… loud ones. Thousands of replies, each one a copy-paste of something that looks human but isn’t. The moderation team responds by killing every post from those bots. But here’s the twist: the bots keep coming, and the [dead] tags pile up. Now the conversation looks like a ghost town. Users complain. They say, ‘It’s worse than Reddit.’ And they’re right.
We’ve seen this before. Reddit’s spam problem. Twitter’s bot armies. But what’s happening now is different. AI doesn’t have to be intelligent to win; it just has to be relentless. One human moderator can delete a hundred bot posts per hour. But a single cheap script can generate a thousand. The math is simple. The math is terrifying.
I watched a thread last night—a thoughtful discussion about Rust’s memory safety. Ten replies. Five were [dead]. The remaining five were people asking, ‘What happened here?’ The original human conversation was buried under the absence of conversation. The forum didn’t look active; it looked like a museum of deleted attempts.
Here’s the uncomfortable truth: the platforms we love are choosing between two bad options. Option A: Let the bots run wild, and the site becomes a sewer of AI-generated mediocrity. Option B: Kill the bots aggressively, and the site becomes a barren wasteland of [dead] badges. The internet is now a binary choice between being flooded with garbage or looking like a ghost town. There is no third option—not yet.
I’ve seen firsthand what happens when a community loses its signal. I was part of a forum that prided itself on ‘no dumb questions.’ Then the AI bots came. No one trusted the new members. Quality collapsed. The founders shut it down in six months. They said it was ‘too much work.’ What they meant was: the cost of defense exceeded the value of the community.
This isn’t a technical problem. It’s a structural one. Anonymous, open internet forums were built on a fragile trust that people are who they say they are. AI has broken that trust, and we can’t glue it back together. We are now witnessing the functional death of a model that powered the web for two decades. The golden age of the open internet forum is over. And the only people who seem to notice are the ones staring at a screen full of [dead].
FAQ
Q: Isn't this just a temporary problem that better AI detection will solve?
A: Better detection helps, but it's an arms race. Every detection upgrade triggers a new generation of bots. The real issue is that the cost of generating spam is now near zero, so the supply is infinite. Moderation can never catch up—it's a losing battle.
Q: What practical change can users make right now?
A: Stop treating anonymous posts as trustworthy. Demand identity signals—like verified accounts, reputation systems, or proof of work. Until platforms require some form of cost to post, the bots will always win. Also, if you see a [dead] thread, don't ignore it—that's the signal of a system under siege.
Q: Isn't the contrarian view that AI bots actually improve forums by generating more content?
A: Quality matters more than quantity. Bots create noise, not signal. A thousand fake replies drown out one genuine expert. The contrarian take is that we should embrace smaller, curated communities with high barriers to entry—like private Slack groups or paid membership forums. That's not the open internet, but it's the only way to preserve quality.