Read Time: 4 minutes 14 seconds

Artificial Intelligence isn’t coming. It’s here.

Just six months ago, OpenAi launched ChaptGPT to the public. And what started with a flurry has quickly cascaded into a full blizzard.

According to Zain Kahn, editor of the AI Newsletter Superhuman, more than 1,000 AI tools were released in April alone.

And just like me, I bet you’re wondering how to ski this new slope safely despite the avalanche danger. I mean, a new tool every hour is a lot to keep up with.

Here’s the playbook I’m running:

For today, let's get started with the help of some experts.

First, I think it’s important to put these tools in the right context.

According to OpenAI CEO Sam Altman, "The right way to think of the models that we create is a reasoning engine, not a fact database. They can also act as a fact database, but that's not really what's special about them – what we want them to do is something closer to the ability to reason, not to memorize."

This distinction was unlocking for me. While we can get some lift from using them as super search engines, the real power comes from working with them like assistants or sparring partners. Except these assistants never tire and have been trained on nearly all written knowledge through 2021.

A word of caution: Because it assists you through reason and not exclusively facts, it can lie to you. So before you take an output at face value, it might well be worth double-checking depending on your use.

Now let’s get started.

For today, I’d like to show you a few ways to get an immediate ROI from using ChatGPT. If there’s interest, we can delve even deeper into future issues or explore other AI productivity tools.

I found the most useful basic prompts help accelerate learning. Where Google makes the most likely information available, ChatGPT provides the synthesis.

That’s what you’ll get with basic prompts.

For example, every addition of Superhuman includes a sample prompt to help you experiment with ChatGPT.

I’ve modified a couple to showcase how they can be used for management.

One additional tip. Don’t be afraid to regenerate a response, even without adjusting the prompt. The variations can be pretty wild, and sometimes, the second time is the charm.

Finally, Zain shared a few more prompts yesterday that are simple in design but powerful in their results.

Perhaps the most leveraging insight to move from using ChaptGPT as a super search engine to a research assistant is realizing you can ask ChatGPT how to use ChatGPT.

While this might seem meta, you’ll quickly see that these “trees” create huge cascades of value. Aadit Sheth, also known as “Mr. Prompts,” shared this in a recent newsletter. He’s going to help us out with taking our prompts up a level of sophistication.

The next intermediate step to take is realizing that the output doesn’t just have to be a text response but can actually be full-fledged draft deliverables.

For example, let’s say you need to hire someone. There’s a lot that goes into building a credible recruiting process and avoiding a bad hire. Imagine having an experienced recruiting partner you could give basic notes to, and they’d produce a job description, identify sourcing strategies, write interview guides with grading rubrics, etc.

ChatGPT is that assistant:

One additional tip: You can ask for output in various formats. You’ll notice in the above I specified a table. Don’t be afraid to as for plans, bulleted lists, or elaborate prose. You can also have it channel specific people past and present.

Speaking of…

Now it’s time to really put ChatGPT to the test. For this section, I called on Alex Banks from Though the Noise.

I asked you all and my community of 125,000+ leaders on LinkedIn and Twitter what parts of their job took the most time with the smallest return. Those seemed like areas ripe for AI disruption.

I handed those over to Alex. Here’s how he decided to attack the challenge.

The results back were outstanding. A mix of resources, frameworks, practical advice, and prompts to get even deeper with my ChatGPT assistant’s help.

Here’s the example playbook it generated for better meetings (you can stack it up against my advice for 1-on-1 meetings from a few months back):

What’s powerful is how extensible this prompt framework is.

Persona - Can’t imagine working with Ray Dalio? Drop in Satya Nadella or Sheryl Sandberg. Pick a public persona closest to the person you’re partnering with.

Problems List - While mine was sourced from a broad community, you can replace it with the discrete problems on your team.

Solution - While I got resources, frameworks, specific context, and further prompts, you can ask for anything. Success or failure stories. Drafted emails.

As a thank you to our experts, I wanted to make it easy for you to subscribe to their newsletters. Click any of the links below to signup in one click:

Through the Noise by Alex Banks
“AI. Without the noise.”

Neat Prompts by Aadit Sheth
“AI prompts and tools to become a marketing and productivity machine.”

Superhuman by Zain Kahn
“Learn how to leverage AI to boost your productivity and accelerate your career.”

We want the MGMT Playbook to help you win. This week we detoured off our typical path and into AI. Tell us that was a good idea.

Our mission is to impact 1,000,000 leaders positively. If this playbook would help someone you know start leveraging the power of AI to lead, please forward it along.

And if someone sent this edition to you, please don't leave without hitting that Subscribe button now.

Thank you for reading. Appreciate you!

Dave

PS - Last week’s edition on Turning Your Team Into A Trust Factory generated a lot of chatter. Check it out to see what you missed.