Balancing the Duality of Generative AI

I recently came back from holidays and a welcome break from the daily hype of Generative AI.

And it occurred to me that beneath the promises of what AI tools can do, there's a duality that underpins every aspect of artificial intelligence.

Each part of that duality has positive and negative consequences, and the two extremes seem to be in a constant battle to gain the upper hand.

Which means that every time you use a Generative AI tool, you experience the conflict first hand. And the push and pull, that jostling for position, may be responsible for some of our uncertainty and stress.

Here's a chart to show you what I mean.

On the left, you have positive aspects from the work marketers and comms pros produce. On the right, the negatives.

Because Generative AI is grounded in language and communications, it has a similar range.

Its plusses include an ability to quickly come up with an endless stream of ideas (including many stinkers), edit text and images, produce visual concepts, have conversations and provide alternate points of view. It's also a good platform for A/B testing.

But the negative side, if left unchecked, could lead to catastrophes like blatant errors, mass persuasion, disinformation at a scale we've never before seen, biases, and an inability to explain its thinking or answer why.

And while many of today's AI outputs are less than stellar, its overall performance is improving in leaps and bounds, in part because we're training it with our data every time we use it.

So How Do You Strike a Balance?

I wish I had a good answer. But it's an individual choice you'll want to make.

One thing you can do is consider the +/- chart as a dashboard, where you adjust the settings to each of the parameters when you use AI, or as a rationale for the times you don't.

Let's use this newsletter as an example. I like writing! So while AI may be faster, I still have a big enough ego to believe I'm a better writer than a machine. So I'll strive for originality over cliche and that means most of the heavy lifting falls to me.

Creativity is another place I'm not willing to make tradeoffs so I would take on most of the responsibility and not rely on AI. I would also dial up safety.

But as we move down the list, my settings become more nuanced. For instance, I'm not as concerned with my own privacy on social and digital media and would dial up the public availability of my content.

For authentic vs. synthetic, I might set it closer to the middle as I often use text to image tools for visuals. But I would also disclose.

And depending on whether the topic I was writing about was more speculative or based on facts, I would increase either feeling or data.

These are just a few examples. But they give you a place to begin.

Get to Know the Frontier Models

It's also a good idea to test out the many chatbot models to figure out which one performs best for your various tasks.

Are you a ChatGPT purist? Do you prefer Google's Gemini, the more serious, if chatty Claude from Anthropic or Microsoft Copilot?

Or does your rebellious streak draw you to X's Grok?

Every large language model, or LLM, has its own personality and quirks. And it's a good idea to determine the strengths and limitations of each.

In this week's Digital Marketing Trends video, I provide an overview to get you started. Check it out and let me know what you think.

Data Science for Marcom Pros

LLMs also share another common aspect: data. The systems consume data ravenously and use it as the foundation for their statistical predictions.

So start thinking of your work as data generation (not just writing, producing videos or design) to gain a perspective on the types of data you use and how they're processed by Generative AI apps.

That's what Deirdre Breakenridge and I will be talking about on the next #DM Show with our special guest Alex Sévigny, PhD, APR , on Friday, July 19 at 1:00 pm ET.

Alex is a professor at the McMaster University Master of Communications Management program and its past director. He’s also an author, data scientist and strategic digital advisor to many public and private-sector organizations.

Alex will talk how marketing and communications professionals can apply data science principles to your strategy, content and campaigns by quantifying media narratives with natural language processing or using data to spot emerging trends.

It's a free #LinkedInLive #officehours event. To RSVP, click this link or the photo below.

Follow Me on LinkedIn

No ifs, ands or buts, you've come to the end of issue 104.

Before I go, I want to share some news.

I recently did an interview with Vianca Meyer for the DesignRush podcast. The subject? Generative AI opportunities and ethical watchouts . Have a listen and let me know what you think.

And I was thrilled to receive the 2024 Canadian Public Relations Society Thought Leader award . As a longtime member and supporter of CPRS, that meant a lot and I want to thank them for the honor.

Thank you to all of you who follow me and subscribe, read, comment and share this newsletter! It's great having all of you along for the ride.

This newsletter comes out every two weeks. But between issues, I share shorter daily posts with my take on digital marketing and the latest on generative AI. It's another way to stay on top of the trends.

And while you're at it, follow the Future of Marketing Institute , too. Every day we post content and perspectives on where we may be heading and what the shift might mean.

Let me know if you have questions about any of the videos in Digital Marketing Trends, or my other LinkedIn Learning courses . You can also visit my website and send a message or a question.

How do you find a balance with the duality of AI? Please share your thoughts in the comments below.

There's no duality involved in whether I'm coming back with another issue. See you in a couple of weeks!

Note: All the content in this post was written by a human—me and not Martin-bot.

For another perspective on the duality of AI , here's a post by Jon Tan.