In this insightful article titled “Making AI Work: Leadership, Lab, and Crowd” on One Useful Thing, Ethan Mollick unpacks a crucial dilemma facing organizations today: while AI is supercharging individual productivity, most companies aren’t seeing these gains reflected at scale. He explores why this disconnect exists and introduces a practical framework—Leadership, Lab, and Crowd—to help organizations unlock AI’s true potential and transform the way work gets done.
Key Takeaways
- AI’s Promise Remains Untapped at Scale: Despite clear productivity boosts for individuals, most organizations struggle to translate these into company-wide improvements.
- Leadership Sets the Vision: Leaders must go beyond generic memos and paint a vivid, motivating picture of an AI-enabled future—one that inspires teams to experiment and innovate.
- The “Lab” Drives Experimentation: Establishing a cross-functional “Lab” empowers teams to rapidly test, benchmark, and scale AI solutions tailored to real business needs.
- The “Crowd” Powers Innovation: Frontline employees often lead the way in discovering creative AI uses. Encouraging open experimentation and sharing best practices helps spread these gains.
- Overcoming Hidden Barriers: Many employees are “secret cyborgs,” quietly using AI but hesitant to share learnings due to compliance fears or lack of support. Leaders must foster a culture of safe, transparent experimentation.
- Learning is the Real Challenge: Adopting AI isn’t just about new tools—it’s about learning, iterating, and rethinking how work happens at every level.
- Practical, People-Centric Approach: The Leadership-Lab-Crowd model offers a clear, actionable path for organizations to move from isolated AI wins to widespread, sustainable transformation.
Read the full article here: Making AI Work: Leadership, Lab, and Crowd by Ethan Mollick
What’s your experience with AI at work? Are you seeing real change, or do you feel stuck in the “AI productivity paradox”? Share your thoughts below!
Just Being Resourceful! 
Leave a comment