By Theresa Hegel and C.J. Mittica

C.J. Mittica: Theresa, there’s been a ton of discussion, speculation and even fear mongering since OpenAI released ChatGPT late last year. Reactions seem to range from “This can help your business” to “This is going to put people out of a job and might be the downfall of the human species.” We just published a feature that you wrote covering many different business use cases for generative AI, and in the course of reporting that article, you talked to many experts both inside and outside the promo industry. After all your conversations and research, where do you fall on ChatGPT? Do you believe we should embrace AI – as our CEO Tim Andrews has done – and promotional products businesses should dive right in? Or should we all be treading very carefully?

Theresa Hegel: I think ChatGPT and other generative AI can be a helpful supplemental tool for knowledge workers. In the course of reporting my story, I signed up for an account for the free version of ChatGPT, and I’ve experimented with it here and there. I’ve used it to generate story ideas on a variety of topics. I’ve also asked it to write full articles just to see what it would come up with. And it spits out some serviceable, if uninspired, prose (on the level of a five-paragraph essay written by a college freshman taking basic composition). I would never copy and paste anything it provides because it’s just synthesizing things it finds on the internet, not creating anything new. So, there are plagiarism and accuracy concerns. Plus, I’m kind of a snob about my own ability to write well. However, I think people who take some time to learn how to prompt the AI properly can use it to boost productivity and efficiency in a variety of areas.

In its current form, I don’t think it’s a replacement for human workers and definitely doesn’t live up to the all the hype it’s getting. Think of it is just another tool in your digital toolbox. What about you? Have you tried AI at all?

CM: I’ll come clean and admit I haven’t tried it yet, though I’ve definitely been curious, especially when I have to transcribe interviews (which, as you can surely attest to, is one of the more thankless tasks in our profession). Our editorial team has used it for a couple “experiments” with our content just to see what the results would look like, including taking a crack at a Counselor Top 40 profile. As you said, the writing was serviceable and sounded authoritative, but was either wildly off-base or saying something without really saying anything at all.

To me, I currently view generative AI as a well-meaning intern: somebody who can do decent work but needs a firm hand and lots of guidance to get the best results. But we also know interns can learn quickly and become all-star employees. It’s no different with AI. What kind of growth can we see – and how soon might it happen?

TH: I think the field is going to grow quickly, and we’ll see more and more businesses incorporate AI into their models and marketing. AI is hot, and investors are throwing cash at it: The New York Times recently reported that venture capitalists have spent an estimated $1.37 billion on AI-related deals in 2022. So, we’re definitely going to be seeing more AI-powered tech going forward, but one of the experts I spoke with for my article pointed to the Gartner Hype Cycle and noted that we’re currently at the “peak of inflated expectations,” but that many of us are going to end up in the so-called “trough of disillusionment” as the limits of artificial intelligence become more apparent. Eventually, things will level off and we’ll hit the “plateau of productivity.” We’ll embrace the efficiency gains this tech can offer without fearing that the robot takeover is imminent.

“In its current form, I don’t think AI is a replacement for human workers and definitely doesn’t live up to the all the hype it’s getting. Think of it is just another tool in your digital toolbox.”Theresa Hegel, ASI Media executive editor, digital content

I’m interested to see what happens in terms of ownership of AI-generated content. There are a lot of questions surrounding this area, especially since many of these AI models were trained by scraping images and other content from the web without seeking any kind of permission first. Image-generating AI platforms are already in the midst of legal battles, with artists saying their rights have been violated since their images were fed to the AI without consent, compensation or credit. And I expect there could be similar lawsuits coming in the future related to written content. Increased litigation and oversight could definitely put a damper on AI advancement, but is that a bad thing? Or an important step in the process?

CM: Interesting questions. I want all the benefits of AI but don’t want (or at least don’t want to know about) the murky lines that need to be crossed to get there. Facebook at least had the courtesy to connect you with friends in exchange for your data, even if not everybody realized the deal they signed up for.

Actually, I think that cuts right to one of the core issues with AI: transparency and authenticity. For customer service – and not everyone agrees with this – I’m willing to interact with a chatbot online to get information quickly. I just want to know it’s AI. Likewise, I’m probably not the only one who was amazed by an AI Morgan Freeman voice or had a good laugh at a completely redubbed famous movie scene.

But the potential of deep fakes troubles me. What’s real and what’s not? That’s why I think it’s important for promo businesses in certain instances to disclose when they’re using AI. And going beyond that, I believe there’s going to be a renewed demand for in-person encounters and face-to-face dealings. “Real” is going to be a very valuable commodity.

“I believe there’s going to be a renewed demand for in-person encounters and face-to-face dealings. ‘Real’ is going to be a very valuable commodity.”C.J. Mittica, editor-in-chief

TH: I agree with all of this. Many of us appreciate the convenience that chatbots can bring, especially when you have a simple question that needs to be answered after hours. But nobody wants to feel duped. I’m also troubled by deep fakes, and it seems that even some of the generative AI platforms are concerned as well. A few weeks ago, the AI program Midjourney shut down free trials of its service, citing “extraordinary demand and trial abuse,” according to the Washington Post. One of the deep fakes that really seemed to spark people’s imagination was an image of the pope in a white puffer coat. It’s a funny image – just close enough to reality that it seems plausible – and it’s a mostly harmless example of this phenomenon. But oftentimes these deep fakes aren’t so benign, and it could really wreak havoc on people’s lives if not checked in some way.

CM: To me, this is world-altering technology that’s still in its infancy. There’s going to be hiccups, controversy, even some very damaging side effects. The arguments of whether AI like ChatGPT is “intelligent” – and what intelligence actually means – are fascinating. The unknowns are daunting, and I can understand if people are hesitant. But to do so might mean missing out on something that can potentially be very helpful for your business.