Alex Lieberman's company 10X runs a "Creator Cup"—a month-long, gamified challenge encouraging employees to post on social media. The primary goal isn't leads, but attracting top engineering talent by showcasing the team's expertise. The ROI is measured in hires, making the prize money a worthwhile recruiting expense.
When mapping a process to apply AI, don't just document current workflows. Instead, design the ideal process you would have with zero constraints (e.g., unlimited time, budget). This approach allows you to leverage AI to automate ideal-state tasks, like research or content repurposing, that are currently impossible due to resource limitations.
To prevent recurring errors, Lieberman's AI system maintains a "Content Lessons" markdown file. When he gives feedback on a draft, the system abstracts the changes into reusable lessons and logs them. The AI then consults this file for all future drafts, creating a powerful reinforcement loop that improves quality over time.
As AI commoditizes technology, traditional business moats weaken. Alex Lieberman argues that building a trusted distribution channel—a media company on top of your business—is a key defensible advantage. This requires empowering employees to become creators and build audience trust at scale, turning the entire company into a marketing engine.
Alex Lieberman's AI company, 10X, uses a variable compensation model for its engineers, similar to a sales team. He believes this unconventional structure self-selects for highly motivated engineers willing to bet on their own abilities. It incentivizes them to stay on the technological frontier, as their earning potential is directly tied to their performance.
After an initial draft is generated, Lieberman's system sends it to an AI "Writer's Council" comprised of personas of famous writers (e.g., David Perel, Morgan Housel). Each persona scores the content from 1-10. If the aggregate score is below a 9, the system automatically triggers a revision loop until the quality threshold is met.
Instead of brainstorming alone, the system uses an AI "interview panel" with personas like Tim Ferriss and Joe Rogan. These AI interviewers ask probing questions about a chosen topic to extract detailed stories, specific examples, and nuanced viewpoints from the user. This turns a solo writing process into a structured, conversational extraction of expertise.
The first step to building an AI workflow is mapping the existing manual process. This exercise alone reveals significant waste and inefficiencies that have nothing to do with AI. The push to adopt AI acts as a forcing function for companies to finally audit and streamline their core operational workflows for the first time.
Alex Lieberman's system starts with an "Oracle" that scans internal tools like Slack and Notion. It analyzes the last seven days of activity to automatically surface and rank potential content ideas, or "spikes," daily. This solves the "blank page problem" by providing a constant stream of inspiration rooted in real work.
Lieberman argues that when his "Content Machine" produces low-quality output, it's not the AI's fault. It's an indictment of the user for not providing sufficiently specific, insightful, or story-driven ideas during the AI interview step. The quality of AI-assisted output is fundamentally limited by the quality of the human input.