AI Writing Without the Slop

Every session from last week's COZORA AI Virtual Summit is available to members right now. That means all the recordings, all the prompts, workflows, and file systems our experts shared, ready for you to implement today. Upgrade to COZORA and you'll also get a live 90-minute workshop every Thursday on Substack. New expert, new skill, every single week.

---

AI Writing Without the Slop

Most people using AI to write are producing content that reads like everyone else’s content. That’s not a tool problem — it’s a process problem.

In this session, I brought in two practitioners who’ve built real systems for solving it: Daniel, who writes about generative AI for everyday users at Why Try AI, and Nick, the creator of the VAST Voice Print Method. Together, they sit at different ends of the AI writing spectrum — and that tension made for one of the most practically useful conversations I’ve hosted on the topic.

The core insight that ran through the entire class: AI defaults to average. It’s trained on everything ever written, most of which is mediocre, and it blends it all together into something inoffensive and forgettable. The writers who are beating this aren’t using better prompts — they’re using better systems. Systems that put constraints on the machine so it can’t retreat to the lowest common denominator.

If you’re a writer, creator, or anyone who publishes regularly and uses AI in any part of that process — or is thinking about it — this session will change how you think about what’s actually possible.

During the class, we:

• Explored why AI writing defaults to generic, forgettable content

• Discussed the full spectrum from AI-averse purists to full AI automation

• Introduced Nick’s VAST framework: Vocabulary, Architecture, Stance, and Tempo

• Learned how voice prints can replace prompts entirely

• Saw Daniel’s Beta Editor skill live, reviewing a real post in real time

• Explored how constraints force AI to produce more distinctive writing

• Learned why copy-pasting popular prompts produces the same output as everyone else

• Discussed how to build a self-correcting voice print that improves over time

• Heard how Nick wrote a 100-day streak of posts almost entirely from his phone

• Answered questions from participants throughout

Implications

The conversation pointed to something more important than any single tactic: the distinction between generic and distinctive is the only line that matters in AI writing — and it predates AI entirely. Slop existed long before ChatGPT, in the form of SEO-stuffed listicles and boilerplate content. AI has simply made it faster and more ubiquitous.

What Daniel and Nick are each building — from opposite ends of the spectrum — are frameworks for encoding distinctiveness before the AI ever touches the work. For Daniel, that means using AI as a rigorous post-production editor rather than a drafter. For Nick, it means documenting his voice so thoroughly that the AI can reconstruct his thinking from a half-formed idea.

The broader implication is this: as AI writing tools become standard, the writers who invest in systematizing their voice now will have a significant and compounding advantage. The tools are commodities. The voice isn’t.