Back/Marketing/Claude/Notion
AdvancedMarketing

Build an AI Content Machine to Beat Writer's Block and Scale Your Brand

A 6-step system, inspired by Alex Lieberman, that uses AI to generate ideas, extract your unique insights through an 'interview', and draft high-quality content that sounds like you, effectively eliminating 'AI slop'.

Build an AI Content Machine to Beat Writer's Block and Scale Your Brand

13:24 to 30:38: Alex Lieberman moves from idea generation to a conversational interview, then turns his own stories and opinions into drafts and derivative formats.

Before you start

What you need

  • A defined audience and content territory
  • Past writing, transcripts, or notes that demonstrate the creator's voice
  • One promising idea plus personal examples and a clear point of view

What you’ll make

A publishable source draft built from the creator's real ideas, examples, and language, plus a short list of derivative pieces.

Tools used

Step by step

The workflow

Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.

6 steps

Step01

The Oracle: Defeat the Blank Page

Connect an AI assistant to your internal tools like [[HIAI:SPAN:4a7FmnpsTr7eGMJy759WpE:Slack]], [[HIAI:SPAN:ZJ7OhN6YiX4rG7NvuOUHjQ:Notion]], and Gmail, as well as external feeds from X and LinkedIn. The system will scan the last seven days of your activity to find and rank 'content spikes', or ideas with strong stories or contrarian viewpoints, delivering a daily list of potential topics.

Example prompt
The system is prompted to scan the last seven days of information from connected internal systems and a pre-defined list of social media accounts, looking for ideas with strong stories, anecdotes, or contrarian points of view.
Step02

The Interview Panel: Extract Your Real Ideas

Select an idea from the ranked list. Instead of writing, have the AI interview you. A panel of AI personas, such as Tim Ferriss or Joe Rogan, will ask probing questions about the topic. Speak your answers aloud, and the system will transcribe them using a tool like [[HIAI:SPAN:4a7FmnpsTr7eGMJy759ZlG:Wispr Flow]] to create the raw material for the draft.

Example prompt
The system uses six distinct AI personas codified as skills (e.g., Tim Ferriss, Barbara Walters) to ask probing questions one by one about a selected topic. The user's spoken answers are transcribed to become the source material.
Step03

Drafting with a Codified Voice

Instruct the AI to draft a post using the interview transcript as the sole source material. For style and tone, provide three markdown files as a guide: a style-guide.md defining your role and goals, a voice-guide.md codifying your writing patterns, and a content-lessons.md logging past mistakes to avoid.

Example prompt
The AI is prompted to use the interview transcript as source material and reference three markdown files (`style-guide.md`, `voice-guide.md`, `content-lessons.md`) to define the tone, style, structure, and persona for the draft.
Step04

The Writer's Council: Automated Quality Control

Submit the draft to an automated 'Writer's Council' of AI personas modeled on writers like David Perell and Morgan Housel. Each persona scores the draft on a 1 to 10 scale and provides feedback. If the aggregate score is below a 9 out of 10, the system automatically revises the draft using the feedback and resubmits it until the quality threshold is met.

Example prompt
Six AI personas modeled after great writers review and score the draft. If the aggregate score is below a 9 out of 10, the system automatically uses the council's feedback to revise the draft and resubmits it.
Step05

The Lessons Loop: Create a Reinforcement System

After the AI draft is approved, perform your final manual edits and publish the content. The system then runs a 'diff', comparing the AI's last version with your final published piece. It extracts general lessons from your changes and asks for your approval to add them to the content-lessons.md file, improving future drafts.

Example prompt
The system asks the user: 'Do you want these lessons to be added to the lessons file?' If the user says yes, the feedback is permanently logged in the `content-lessons.md` file.
Step06

Repurpose and Distribute

With the final, published content as the source, instruct the AI to repurpose it for different platforms. Use a direct command like 'repurpose this into three short tweets and two long LinkedIn posts'. The AI will adapt the content for each format, using the rules defined in your voice guide.

Example prompt
Use a simple, direct command like: 'repurpose this into three short tweets and two long LinkedIn posts'. The AI will adapt the content for each format based on the styles defined in your voice guide.

What good looks like

  • The draft contains claims and stories supplied by the creator
  • The opening, structure, and examples serve one clear argument
  • Repurposed pieces preserve the source idea without repeating identical copy

Build your next product with ChatPRD

Turn an idea into a PRD, user stories, and a plan.

Try ChatPRD free

After the steps

Runbook notes

How to recover when the loop fails and where human judgment helps.

Recover

If it goes sideways

The draft sounds polished but could have been written by anyone
Return to the interview and ask for specific moments, disagreements, phrases, and consequences.
Repurposed posts flatten the original argument
Give each format one distinct angle, audience, and purpose while preserving the source position.

Start shipping
better products.

Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.

Free to start
No credit card
SOC 2 certified
Enterprise ready