Ted Sarandos is Right. Start Using AI Today to Stay Competitive

This short article is for anyone who isn't yet convinced that AI Large Language Models (LLMs) such as ChatGPT are extremely useful in their daily personal and professional life. It might be that you already gave one of these AI models a quick trial run and were not impressed, or it might be that you haven’t yet taken the time to give these latest AI innovations a try. If this describes you, I'm here to tell you that you are missing the boat and falling behind. As Netflix CEO Ted Sarandos said in a recent interview, “AI won’t take your job. The person who knows how to use AI well might take your job.” The good news is that it’s absolutely not too late to get up to speed, and it’s not at all difficult to ramp up and understand how to access the capabilities of this amazing new technology to make you more effective and efficient in your daily life. I’ve got 7 simple quick tips to get you rolling. So let’s get started.

Tip #1: Bring AI into your life starting now!

Get into it! The only way to learn how this transformational technology can help you be more effective and productive—what it’s great at and what it’s not so good at—is to start using it. AI is like having your own personal executive assistant who happens to be super smart, super fast, super organized, and who just happens to have deep knowledge about the exact topic or challenge you are facing. The LLM can proofread your emails and other written work, brainstorm ideas for a trip, a dinner, or a party—whatever. You’ll be amazed. Get your AI reps in NOW with everyday simple queries so you can gain the necessary comfort and expertise to be ready to put this amazing technology to work for you when you really need it!

Tip #2: Be sure you are using the latest version of your AI model of choice.

If you’re going to evaluate a technology that is evolving exceedingly rapidly, it only makes sense to test the latest available version of that technology. The model I recommend, ChatGPT-4o, went into wide release in May 2024. Compared to its predecessor, it is trained on vastly greater amounts of data and has significantly more processing power. This makes the latest version much more conversational and capable of working with nuance—understanding what you need and providing contextually relevant results that are more precisely calibrated to those exact needs. It’s also multimodal, meaning it can work seamlessly with text, imagery, and voice. As part of your prompt, you can upload an image of an object or a screenshot of a spreadsheet you are working with, and the model can interpret that picture or “read” that spreadsheet. This can be extremely useful and dramatically expands the model’s capability. ChatGPT 4o can generate picture quality imagery based on your personal instruction. You can also talk to it, and it will respond in a conversational manner that is very natural. In the world of AI, using an outdated version is like carrying a flip phone.

The next four tips are about effective prompting. A prompt is the term used for the input text or question that the user puts into the model. Essentially, it’s the user’s way of asking ChatGPT to perform a task, answer a question, or engage in a conversation. The quality and clarity of the prompt can significantly influence the relevance and accuracy of the response generated by the model. Mastering the prompt is the art of using AI, and it takes a bit of practice to learn what info the AI model needs in the prompt in order to give you the quality response you are looking for. The following prompt tips are based on my personal experience and reading I’ve done on the topic.

Tip #3: Your prompt should assign AI a role.

No one is exactly sure why this works, but it has proven to be very effective. Start your prompt by telling the model a role you want it to assume. For example, depending on the question:

• You will act as my social media copywriter.
• You will act as my proofreader.
• You will act as my strategic business advisor.
• You will act as my Human Resources expert.
• You will act as my therapist.

Tip #4: Your prompt should provide a clear, detailed, accurate description of exactly what you need.

The more detailed and descriptive you are regarding exactly what you need, the better the AI result will be. For example:

• Bad prompt: “Give me 10 clever headlines for an article about public speaking.”
• Better prompt: “Act as my social media copywriter (assign it a role from tip #3 above) and give me 10 clever, intriguing, and humorous headlines for the following article I will publish on LinkedIn to engage potential customers for my online public speaking course. My goal is to maximize engagement with LinkedIn users who might decide to purchase my services as a result of reading the article.” (Then copy and paste the full article into the prompt box). The model will “read” your article and give you the list of 10 headlines in seconds.

Tip #5: When prompting, create rules that provide explicit clarity as to what the output should be.

Don’t worry about micromanaging; the LLM won’t be offended! The more explicit you are, the more likely you are to get the output you need.

• Bad prompt: “What is calculus and how is it used?”
• Better prompt: “Act as a teacher and explain what calculus is used for in the real world of 2024. Your response should be approximately 500 words and at a 12th-grade level.”

Tip #6: Iterate as necessary.

The model’s first response might not be exactly what you are looking for, so follow up with a new prompt. In the headline example above, you might find that based on the initial response, none of the 10 headlines are quite right, but #6 is the closest to what you’re looking for. So you might follow up with the following prompt: “Give me 5 variations of headline #6 on your list. These variations should focus on the idea that the public speaking course is super flexible and convenient for busy people.” You can continue to follow up on the model's responses, iterating until you get exactly what you need.

Final Tip: Cross-check to confirm the accuracy of the response you get from AI before acting on it or quoting it.

Just like you would for an internet search or information you acquire from any source, you want to do your due diligence to confirm that the response the AI model has provided is accurate before you act on it—especially for critical tasks. There’s a lot of bad info on the internet, and AI is trained using vast amounts of information from the internet. So confirm the response from a separate source. You could also reply with the prompt: “Cite the source for the information you provided in that response.”

AI LLMs are an innovation on par with the invention of the internet. You don't want to be left behind with this game changing technology. Follow these tips, and you’ll find yourself increasing the speed and quality of your output by a factor of 3-5X. Good luck and have fun!