Six Need-to-Know ChatGPT Insights and Implications

Welcome to Leading Disruption, a weekly letter about disruptive leadership in a transforming world. Every week we’ll discover how the best leaders set strategy, build culture, and manage uncertainty all in service of driving disruptive, transformative growth. For more insights like these, join my private email list.

ChatGPT burst onto the scene in late November 2022. And it’s taken the technology world by storm.

For something still in beta, it’s been adopted astonishingly quickly and widely reiterated and integrated into existing technologies.

In just over two months, ChatGPT reached 100 million active users. To put that into perspective, it took TikTok nine months to reach that level. Instagram needed two and a half years to hit 100 million users.?

This rapid change has brought up lots of questions and even more concerns. And ChatGPT just announced its newest version GPT-4 on Wednesday this week. So much to keep up on!

On Tuesday’s livestream, I shared six things you need to know about ChatGPT and other AI tools. (Don’t worry — it’s not all doom and gloom!)

1. People are very aware of AI.

AI is something people are paying attention to — and they’re concerned about it. They don’t understand it. They think, “Wow, there’s a lot of jargon.”?

They’re concerned about the impacts on their jobs, especially if they create content. Creatives worry about the implications for their art and their ability to be compensated for their work. There are certainly copyright and IP ownership issues: If you use generative AI to create art, who owns the art?!

At the end of the day, the technology and its capabilities aren’t explained very well, so people fundamentally don’t trust it.?

2. ChatGPT has limitations.

AI is not omnipotent! It doesn’t have all the answers! Keep in mind that:

• It’s a limited database. It only dates back to 2021, so it’s not the best for real-time data.?
• It doesn’t always give you the right answers because it often doesn’t understand your context or there’s some ambiguity in your question.
• It doesn’t include citations or references, so you might not have full confidence in the answer.
• It only does text. There are no visuals — no photos or videos — built in. Yet. (Note: the next iteration GPT-4 does have image description.)

3. There are many untapped areas.

Most companies don’t have the capacity to look at huge amounts of internally generated content. They don’t have the time or resources to “find the needle in the haystack” — often because they don’t even know what they’re looking for.?

But we could use AI to query databases and make better use of all that data we’ve gathered.?

Let’s say you’re looking at search engine content. You want to determine the keywords you should use in your next ad. Right now, there’s way too much data to plow through. AI can sift through the information and say, “Hey, change this. Shift your ad buys. Change your copy.”?

4. It will require new skills.

One of the most interesting implications is AI’s ability to enhance our creativity. I use it for initial drafts or when I’m stuck because it helps me plow through roadblocks and inject new ideas into the content I’m creating.?

But to open up those possibilities, you have to ask the right questions and be able to refine your queries. When you change the permutations (even slightly), suddenly you’re increasing the potential content you can tap into.

I’m reminded of how, over the last 20 years, we all learned to use search really well. Now we’ll need to acquire new skills — formulating questions, developing knowledge and expertise, applying curiosity and creativity — to use generative AI successfully.

5. And it will put a premium on “made by humans.”

There will always be a premium on human creativity and insightfulness. Can you create helpful, reliable content that puts people first with AI? Yes, but it’s the additional information, the personal experience, the expertise and authority that will become a quality marker.

Sure, ChatGPT can help you create a first draft, and it can be great. It can help you better understand different perspectives, remove biases, and pull from the wisdom of crowds. It can augment the experience we bring to a particular problem, but it will always be missing that human touch.?

“Made by humans” — even if its assisted by AI tools — will become a new mark of pride and a way to build trust.?

6. But we need to be mindful consumers.

ChatGPT has some major issues, most notably what the creators call “hallucinations.” It goes down rabbit holes and creates echo chambers. (For an excellent example of this, check out Kevin Roose 's adventures!)

It can get stuck because it’s trying to give you what you want. The more you engage, the easier it is to get stuck in a rut. And, depending on how you’re using AI, it can create homogenous, unoriginal content that decreases diversity — and increases your own private echo chamber.

This is a huge wake-up call. If we want to incorporate AI technology into our businesses, we need to be very mindful of how we do it. We need to be sensitive to the fact that technology is moving faster than our customers — and, frankly, faster than we can establish a level of trust.?But there are opportunities here: to play, to experiment, to think about and understand the impacts, to increase productivity, to drive creativity, to do better work. We just have to harness its power — and bring our customers along on the journey.

And next week I’ll share some ideas on how we can build that trust. I hope you’ll join me for another lively conversation on Tuesday, March 21 at 9 am PT / 12 pm ET.?

Your Turn

I’d love to hear your experience with ChatGPT! What most excites you about it? What impresses you? What worries you? I’m curious, as someone knowledgeable about technology, disruption, and business, how you’re thinking about and approaching new generative AI tools like ChatGPT. If you’ve had a chance to play around with it, please share your thoughts in the comments below!