The Four Rules of AI: Essential Insights from a Must-Read Book

As avid readers and AI enthusiasts, we find ourselves navigating an intricate jungle of newly published books in the topic of AI. With so many titles flooding the market, it can be a demanding task to sift through them all and discover which ones truly offer practical insights and actionable findings. But one definitely stands out: Ethan Mollick's "Co-Intelligence: Living and Working with AI," which not only demystifies the complexities of AI but also provides practical frameworks and strategies for integrating AI into various aspects of work and life.

One standout takeaway is his Four Essential Rules for Working with AI (plus my fifth one at the end).

Let's explore these four rules together:

1?? Always Invite AI to the Table: You won’t know what AI can or can’t do for your role or industry until you try. The only way to uncover its potential is through consistent, hands-on experimentation—use it as much as possible.

Keep also in mind the concept of the Jagged Frontier: AI excels at certain tasks but falls short in others. Your job is to understand these strengths and weaknesses and navigate this uneven landscape. Keep experimenting and refining along the way. And to effectively do that, always invite AI to the table!

2?? Be the Human in the Loop: While AI can outperform people in specific tasks, it can’t handle the entire job. Focus on your strengths—these are the areas where you’ll continue to outshine AI.

Remember: AI works best when humans remain actively involved, ensuring oversight, judgment, and creativity where machines fall short. The synergy between human expertise and AI is what drives the best outcomes.

For instance, if you develop an AI feature that generates recommendations for the next best step in a procurement process, ensure that humans review these suggestions before full automation. Implement a feedback mechanism to keep humans involved throughout the process.

3?? Treat AI like a Human: AI models are trained and refined using human language, which means they respond best when given instructions as if they were a person. Assign tasks clearly, just as you would to a colleague—it often helps AI produce more relevant results.

However, context is crucial. Without it, AI may default to generic outputs. Use personas to guide it: for example, “You are an expert marketing manager in Europe focused on technology ventures working with the US.” This framing puts the AI in the right mindset and generates better responses than vague or context-free prompts.

A quick funny remark highlighted in the book: if you tell AI to "take a deep breath" (just like a human!) in your prompts, it might just respond with a more thoughtful and detailed answer!

4?? Assume This is the Worst AI You Will Ever Use: We are still in the early stages of AI development, and the technology is evolving rapidly. What you’re working with today will likely be the least capable version you’ll ever encounter.

Remember Will Smith’s AI-generated video of him awkwardly eating spaghetti from just a few years ago? That was an early example of text-to-video applications—and look how much those tools have improved since then!

This is just one example, but it offers a practical glimpse into how quickly AI capabilities are evolving and what’s possible in the near future.

My Bonus Rule: Develop an Insatiable curiosity around AI:

Stay curious and eager to learn about the evolving capabilities of AI. Explore its potential applications and keep experimenting to discover how it can enhance your work and life.

Embrace curiosity, be open to experimentation, and create a sandbox for testing ideas. Share tips and tricks with your colleagues and friends, and most importantly, have fun with it! When applied correctly, AI can give you a significant edge!

And don't forget to check the book below (link under the picture)!

#AI #Innovation #EthicalAI #Leadership #GenAI #Procurement