AI slop is killing companies.
In the last 2 years we’ve taken the cost of content down to zero, so the supply of content has gone through the roof.
There’s not time to read it all. So we don’t. We complain about it, but we think ther’es nothing we can do.
Not anymore! I’ve been staring at the problem for months, and I finally realized there’s a way to use AI to fight slop here. A really sharp prompt, thoughtfully deployed, can vastly simplify the review process for artifacts at work, turning an impossible pipeline of content into a self-policing feedback loop.
Imagine if instead of wondering if an AI has written a good or bad Product Requirements Document and having to read it line by line you could get a quick read and feedback instantly from a prompt! Same with blog posts. Same with customer emails. You get the idea. It’s going to save your team DAYS.
Here’s the full list of anti-slop editor prompts I built:
The High-Volume Explosion (10-100x increase):
Blog posts and articles
Social media content (LinkedIn, Twitter, etc.)
Email campaigns and sequences
Sales outreach (cold emails)
Ad copy (Google, Meta, LinkedIn)
Support responses
Product descriptions
Meeting summaries
Video scripts
SEO content briefs
Landing page copy
Follow-up emails
Case study drafts
Email newsletter content
Internal communications (Slack responses, etc.)
The High-Impact Category (lower volume but expensive when bad):
16. PRDs and feature specs
17. Technical documentation
18. Executive memos
19. Client deliverables and reports
20. Proposals and pitch decks
Quite a list! Look, the point is NOT to stop human eyes or human reviews.
The point is to make human reviews matter more.
Andrej Karpathy said something that’s stuck in my mind for months:“99.9% of attention is about to be LLM attention, not human attention” (March 12, 2025 tweet).
He’s right. And I think we need to start building systems that assume that the human attention needs to be on the correct 0.1% (or 1% if we’re aggressive) of work.
That’s what this prompt system is designed to do for you. Put your human attention where it matters most. Incidentally, by doing this you get all sorts of benefits!
As a writer, you learn to write better, because you have a clean feedback loop from prompt to artifact to specific, actionable feedback
As a professional, the AI learns to work with you better (especially systems with memory) as you use this system
As a manager, your team can save you work by doing first round themselves
As leader, you can save your teams days of work and increase quality with this system
As an agentic systems builder, you get some seed prompts you can use to construct automated pipelines to tackle this stuff
It all rests on having an excellent quality gate, and that’s what I’ve focused on here.
Have fun, and go kill some slop today :)
PS. If you’re wondering if I used a prompt to check my work and help with this article, the answer is yes! Working with a prompt improved my writing significantly.
Listen to this episode with a 7-day free trial
Subscribe to Nate’s Substack to listen to this post and get 7 days of free access to the full post archives.
