“Winning takes care of a lot of mistakes, but you can’t overlook them or sweep them under the rug. You’ve got to be professionals and be the type of team that fixes those mistakes and moves on and improves.” D.K. Metcalf, NFL Wide Receiver

You lead a team who, for the most part, wins. If not, you wouldn’t be where you are. 

But, mistakes are made, which, if not addressed, will cause loss. Compiled losses equal new job search. Worse, your own inner turmoil at not having led at your best.

Currently, you are likely trying to win with generative AI. You and your team are on a learning curve, more likely than not winning despite yourselves. That’s the nature of a learning curve. So is falling behind if you don’t keep up.

How do you outperform others when it comes to Gen AI even if you are still mastering skills?

Clear Focus On Right Skills

In a recent Harvard Business Review article, Wlson and Daugherty argued for three essential skills to greater embrace Gen AI: intelligent interrogation, judgment integration, and reciprocal apprenticing.

I love Harvard Business. I also love the simple. So let me rephrase:

Instruct. Interact. Incorporate. 

Instruct

Be Clear and Specific: Clearly state your question or request with as much detail as necessary. Vague or ambiguous prompts may lead to less accurate or helpful responses.

Provide Context: If your question relates to a specific topic, event, or field, providing background information can help generate a more accurate and relevant answer.

Specify the Desired Format: If you prefer the answer in a particular format—such as a list, essay, code snippet, or step-by-step instructions—mention that in your prompt.

Ask One Question at a Time: Focusing on a single question or topic per prompt helps in generating a clear and concise response without confusion.

Interact

Request Citations or References:

If you need factual or data-driven information, asking for sources allows you to verify the details independently. "Can you provide statistics on global internet usage in 2023 with relevant sources?"

Ask for Multiple Perspectives:

For topics that might have varying opinions or interpretations, requesting different viewpoints can give you a more balanced understanding. "What are the arguments for and against implementing a universal basic income?"

Cross-Check Important Information:

Verify critical information using reputable external sources like official publications, academic journals, or trusted news outlets.

Protect Against Your Own Bias

Our prompts might show a tendency to be biased toward what is most recent, most popular, most traditional, most like our own approaches and so forth.

Incorporate

Be the trainer of your own LLM by incorporating contextual data and knowledge.

Be Specific: Clearly state the context or background information that pertains to your question or topic.

Provide Details: Include any necessary details, such as names, dates, definitions, or explanations, to help me understand the subject better.

Ask Direct Questions: Formulate precise questions that guide me on how to use the provided information.

Use Clear Language: Write in straightforward language to minimize misunderstandings.

Quick Adoption Of New Models

Invest

ChatGPT o1 is insane. Get it. Use it. I was struggling with a code problem using ChatGPT 4 and o1 resolved it immediately.

My business partner has been using o1 for more complex and comprehensive policy creation. 

Here is what o1 said about itself compared to previous models:

“ Enhanced training methods and larger datasets contribute to a deeper understanding of language nuances, enabling more accurate and contextually relevant responses. Optimized algorithms improve processing efficiency, resulting in quicker reply times and a smoother user experience. Additionally, better handling of complex queries and reduced instances of errors or ambiguities make interactions more effective. These improvements collectively enhance both the effectiveness and efficiency of the model compared to its predecessors.”

I still think I could have said that better, but such is the beauty of Gen AI; we are partners, bringing the best versions of ourselves together.

That’s what D.K. Metcalf was alluding to in the earlier quote: winning doesn’t mean we have reached our best, it just means that we won that round. Improving is the only way of continually contending.

‍