Stop Sloppypasta Wants You to Quit Dumping Raw ChatGPT Output on Your Coworkers

There’s a word for what happens when someone copies a 500-word ChatGPT response and pastes it into a Slack thread without reading it first. It’s called “sloppypasta,” and a growing number of developers want it to stop.

Stop Sloppypasta is a new manifesto-style website that hit the Hacker News front page on March 16, 2026, pulling in 170 points and 94 comments in under 10 hours. The site coins a term — sloppypasta — that blends “slop” (the now-standard label for low-quality AI content) with “copypasta” (the internet’s word for thoughtlessly forwarded text). Its argument is simple: pasting unread, unverified LLM output at someone is rude because it offloads all the cognitive work onto the person receiving it.

The timing isn’t random. Just five days earlier, Hacker News officially banned AI-generated and AI-edited comments from its platform, a rule that racked up over 3,000 upvotes in less than 12 hours. The anti-slop movement is picking up real momentum — and Stop Sloppypasta is giving it a vocabulary.

The Three Sloppypasta Offenders

What makes Stop Sloppypasta interesting is that it doesn’t just complain about AI slop in the abstract. It names three specific behavioral patterns that most people will immediately recognize from their own inboxes and group chats.

The Eager Beaver contributes AI-generated text to an ongoing discussion with good intentions. The problem: the output is generic, often irrelevant, and blocks the actual conversation flow. Think of the coworker who pastes a GPT summary into a design review thread that has nothing to do with the specific trade-offs being discussed.

The OrAIcle answers a question by forwarding an unfiltered chatbot response. The site compares this to someone replying “just Google it” — except worse, because the AI answer comes with a veneer of authority that the sender hasn’t actually verified. The recipient now has to do the work of both answering the question and fact-checking the response.

The Ghostwriter presents AI output as their own work without disclosure. This is the most corrosive pattern because it lets the sender borrow credibility while exposing their reputation to whatever hallucinations the model produced. It creates what the site calls a “Boy Who Cried Wolf” dynamic — once people suspect your writing is AI-generated, everything you write becomes suspect.

Why “Effort Asymmetry” Is the Core Problem

The central argument of Stop Sloppypasta comes down to economics. Before LLMs, writing something required effort roughly proportional to the effort needed to read it. A coworker who wrote a thoughtful three-paragraph response probably spent 5-10 minutes thinking about it. That investment signaled: “I considered this carefully enough to be worth your time.”

LLMs broke this balance. Now someone can generate 500 words in 10 seconds without understanding a single sentence. But the effort to read those 500 words hasn’t changed — and the effort to verify their accuracy has actually increased, since you can’t trust the source the way you’d trust a human expert.

This is what Stop Sloppypasta calls “cognitive debt.” Every unverified AI paste forces the recipient to:

Read through typically verbose, over-formatted output
Figure out which parts are relevant and which are filler
Independently verify any factual claims
Determine whether the sender actually endorses the content

The site argues this isn’t a technology problem — it’s a manners problem. “New tools require new manners” is the core thesis.

The Seven Guidelines — and Whether They’re Realistic

Stop Sloppypasta proposes seven rules for responsible AI sharing:

Read the output before sharing it
Verify facts independently
Distill verbose responses to essential points
Disclose AI involvement transparently
Request-only sharing — don’t inject unsolicited AI text
Link-based distribution — share as attachments, not inline dumps
Augment, don’t replace — use AI to accelerate thinking, not substitute for it

The first three rules are hard to argue with. If you haven’t read, verified, and condensed an AI response, you haven’t done any work — you’ve just forwarded noise. The disclosure rule is increasingly becoming a social norm, especially after Hacker News made it an explicit policy.

Rules five and six are more practical and specific. The idea of sharing AI output as a link rather than pasting it inline preserves the flow of a conversation and lets the recipient choose whether to engage. It’s a small UX insight, but a good one.

The seventh rule — “augment, don’t replace” — is the philosophical anchor. Stop Sloppypasta isn’t anti-AI. It’s anti-laziness. The position is that if you’ve actually worked with an AI tool, verified its output, refined it, and formed your own conclusions, then what you’re sharing is your work. You should feel fine sending it. Just note that you used AI assistance.

How This Fits into the Broader Anti-Slop Movement

Stop Sloppypasta isn’t operating in a vacuum. The first half of 2026 has seen a wave of cultural pushback against unchecked AI content:

Hacker News AI comment ban (March 11, 2026): Moderator Daniel Gackle confirmed that while the rule had existed as informal “case law,” it’s now in the official guidelines. The community response was overwhelmingly positive. One user commented: “I much prefer to hear someone’s true voice even if there are a few inaccuracies, so much of a person’s personality is conveyed through their quirks and mistakes.”

Your AI Slop Bores Me: An interactive protest game where players try to fool others into thinking they’re AI or human. It went viral as a Show HN project and turned into a broader movement against low-effort AI content.

RFC 406i — RAGS (Rejection of Artificially Generated Slop): A tongue-in-cheek “internet standard” proposal for identifying and rejecting AI-generated content.

Cory Doctorow’s “Nonconsensual Slopping”: The author and activist coined the term for forcing AI-generated content on people who didn’t ask for it, framing it as a consent issue rather than a quality issue.

What distinguishes Stop Sloppypasta from these other initiatives is its narrow focus. It’s not trying to ban AI writing or shame people for using LLMs. It targets one specific behavior — the unread, unverified copy-paste — and provides concrete guidelines for doing better. That specificity is probably why it resonated on Hacker News, where the audience uses AI tools daily but is increasingly frustrated with how other people use them.

The Hacker News Reaction: Validation and Skepticism

The HN comment thread reveals how deeply this topic has struck a nerve. User czhu12 described receiving “AI generated tickets” at work that “doesn’t match any of our internal product design” — a perfect example of the Eager Beaver pattern. Another commenter, valicord, offered a memorable analogy: “If I’m asking humans, I want to see human responses. I eat fast-food sometimes, but if I served a Big Mac at a sit-down restaurant…”

User Aurornis captured the detection problem: AI prose often feels like “someone took their half-baked idea and asked the LLM to put a veneer of quality writing on top.” The surface looks polished, but there’s no substance underneath.

Not everyone was fully on board. Some commenters argued that the internet was never high-quality to begin with — user madrox contended that “we need better tools as content consumers to filter content” rather than trying to enforce social norms on senders. Others pointed out the irony that the Stop Sloppypasta website itself uses AI-generated design elements, though the creator (namnnumbr) was transparent about this: the essay was human-written while the site design used AI assistance.

The most nuanced critique came from TZubiri, who warned that conflating transparent AI mentions (like citing ChatGPT in a discussion) with undisclosed sloppypasta dumps dilutes the message. It’s a valid point — a researcher who says “I asked GPT-4 about this and here’s what it said, which I’ve verified” is fundamentally different from someone who pastes raw output without context.

What This Means for AI Tool Adoption

Stop Sloppypasta raises an uncomfortable question for companies building AI writing tools: if the best practice is to read, verify, distill, and disclose, how much time does AI actually save?

The answer, according to the manifesto, is that AI should speed up the thinking process, not eliminate it. You can use Claude or ChatGPT to brainstorm, draft, and iterate — but the final output that you share should be something you’ve personally processed and stand behind. The AI is a thinking tool, not a distribution mechanism.

This framing has implications for workplace AI adoption. Companies that encourage employees to “just ask AI” without building a culture of verification and editing are essentially mass-producing sloppypasta. The problem scales: one person’s lazy paste is annoying; an organization where everyone does it drowns in noise that nobody trusts.

For individual users, the guidelines are practical and implementable today. Read before you paste. Verify before you share. Cut the fluff. Say you used AI. These aren’t heroic asks — they’re baseline communication hygiene for a world where generating text is essentially free.

FAQ

What exactly is sloppypasta?
Sloppypasta is verbatim LLM output that’s copy-pasted at someone without being read, verified, or refined by the sender. The term combines “slop” (low-quality AI content) with “copypasta” (thoughtlessly forwarded text). The core issue is that it transfers the full burden of reading and fact-checking from the sender to the recipient.

Is Stop Sloppypasta against using AI tools?
No. The movement explicitly supports using AI as a thinking and drafting tool. Its position is that if you’ve read, verified, and refined the AI output, what you’re sharing is your own work. The problem it targets is the lazy, unread copy-paste — not AI usage itself.

How does Stop Sloppypasta compare to Your AI Slop Bores Me?
Both are part of the broader anti-slop movement in 2026, but they focus on different aspects. Your AI Slop Bores Me is a viral game that highlights the indistinguishability of AI and human writing. Stop Sloppypasta focuses specifically on the etiquette of sharing AI output in conversations and workplaces, offering practical guidelines rather than commentary.

Does Hacker News enforce rules against sloppypasta?
Yes. As of March 11, 2026, Hacker News officially bans AI-generated and AI-edited comments. The guideline states: “Don’t post generated comments or AI-edited comments. HN is for conversation between humans.” A “Flag as AI” feature is being added to help enforce this rule.

What are the key guidelines for sharing AI-generated content responsibly?
Stop Sloppypasta recommends seven practices: read the output before sharing, verify facts independently, distill verbose responses to key points, disclose AI involvement, share only when requested, use links instead of inline pastes, and treat AI as a thinking accelerator rather than a replacement for your own judgment.

Top AI Product