The Ethics of AI, Digital Likenesses, and the Future of Content Creation

Sullivan Finlay 0:00
One of the examples used in this article about Official AI—one of their main clients is the Humphrey Bogart estate. So you can get pictures of Humphrey Bogart in different situations, and brands can use them however they want. And it looks very realistic, as if he’s still alive. And it’s very, very strange to me. It’s most interesting to see, okay, who goes along with this—what creators, actors, comedians, or influencers engage with this kind of thing? Is a fitness influencer more likely to do this than a comedy influencer? Or food versus a movie critic? At some point, is this just a game you have to play, or are there enough people that are kind of like, “Okay, we’re good with this up to a certain extent, but not where it’s kind of replacing what we do”?

Callan Harrington 0:46
You’re listening to That Worked, a show that breaks down the careers of top founders and executives, pulling out key insights that led to their success. I’m your host, Callan Harrington, founder of Flashgrowth, and I couldn’t be more excited that you’re here. Welcome back, everybody. Sullivan, I think this might be the closest we’ve done these back-to-back.

Sullivan Finlay 1:07
Yeah, both in terms of proximity and time elapsed between episodes—and in our relationship.

Callan Harrington 1:12
I said I wasn’t going to do it anymore unless we took it to a deeper level, right? So, let’s jump right in. I’ll kick this off with something fresh. OpenAI released their deep research functionality. Have you heard of this?

Sullivan Finlay 1:27
Not super familiar with it.

Callan Harrington 1:29
Essentially, deep research combs through academic papers on a subject you’re looking for, citing sources and producing a full research output. It actually takes 10, 15, even 20 minutes to generate. It’s not instant like ChatGPT. Right now, you need the pro version to access it, but I found it incredible.

A few things I’m using it for: competitive positioning—seeing who else is out there in my market; ideation—creating my own research reports; and preparing for calls with prospects or partners. The amount of information it pulls is crazy.

Sullivan Finlay 3:07
So you can pull any individual person’s background or past interviews?

Callan Harrington 3:14
Exactly. Plus, it references big research firms like Accenture, McKinsey, EY—anyone doing deep research.

Sullivan Finlay 3:28
Can you ask it to prioritize or exclude certain sources? Like, “Exclude WebMD, give me Johns Hopkins”?

Callan Harrington 3:36
I think so. It asks follow-up questions, like whether you want results for property and casualty insurance or accident and life insurance. It even asks if you want worldwide data or just North America.

Sullivan Finlay 3:58
This is good. You’ll know it’s gotten too good when I ask it for everything Callan Harrington has ever written, and it starts listing your credentials and then says, “His chief fears are…”

Callan Harrington 4:19
Ohio State University, home of the 2024 national champions.

Sullivan Finlay 4:26
That’s when AI has gone too far.

Callan Harrington 4:30
That’s when you know it’s peaked. Anyway, what’s new in your world?

Sullivan Finlay 4:35
I’ve been really interested in the role AI plays in content creation. Official AI, a Seattle startup, just raised $2 million in a seed round. They describe their platform as a way for brands and talent to create AI-generated image and audio content. Essentially, talent uploads a few photos, an AI model is created, and brands can license that likeness for campaigns.

From my understanding, it’s just for photos—not video. But it got me thinking: at what point does this become sad? Could someone upload their likeness and allow AI to create content “in the style of” them? What does that mean for creative control?

Callan Harrington 5:58
That’s a super interesting topic. I get why actors and comedians are upset about AI replacing them. It makes sense—it can get so real at some point. But can AI replicate stand-up comedy? It can analyze timing, joke structures, and audience reactions, but can it create funny material? That’s a different challenge. Humor evolves. What’s funny now isn’t funny in six months.

And how do you replace the human experience? Part of what makes comedy work is how humans relate to each other.

Sullivan Finlay 6:59
Exactly. And one of the biggest clients for Official AI is the Humphrey Bogart estate. You can get AI-generated images of him in different scenarios, and it looks incredibly realistic. So the question is, where do people draw the line? Do influencers and creatives have to engage with this just to stay relevant?

Callan Harrington 7:54
Yeah, and how will people react? I think AI-generated content could push audiences back toward live events. When everything can be faked, the real becomes more valuable.

Sullivan Finlay 8:47
Yeah. Right now, I have more concerns about AI in the creative space than excitement. Not just because of job losses, but because of its impact on originality and collaboration.

Callan Harrington 9:08
That makes sense. Perfect segue into Work in Progress. Want to kick this one off?

Sullivan Finlay 9:14
Yes! So, we’re working on a comedy show together. A mix of improv and sketch. We’re still in the pitch process, but hoping to perform this year.

Callan Harrington 9:41
We’re calling it Corporate Takeover. Inspired by Tech Roast, but focused on Midwest corporate culture.

Sullivan Finlay 10:19
Yeah! We even made a pitch deck, like we were pitching a VC.

Callan Harrington 10:56
Now, onto One Win, One Loss. My win: we officially niched down into insurance and soft-launched a creative agency. Landed two big clients right away.

Sullivan Finlay 12:07
That’s huge. My win: I’ve been prioritizing collaborations with other creators—Lisa Beasley (Corporate Erin), Matt Hendon, and more. It’s creatively rewarding and great socially.

Callan Harrington 13:15
Totally. My loss: I’m doing too much. Reading Buy Back Your Time made me realize I need to delegate more.

Sullivan Finlay 16:02
That’s a great read. My loss: struggling with brand partnerships. I started recording videos to explain concepts instead of relying on back-and-forth emails. It worked!

Callan Harrington 16:35
That’s smart. Now, Spicy Hot Take: “Specialization is overrated. Generalists will dominate the next decade.”

Sullivan Finlay 20:37
That’s interesting. AI is making it easier for people to be generalists.

Callan Harrington 21:28
Yeah, you can niche in industry, but generalize in skills.

Sullivan Finlay 25:40
Totally. Alright, that’s a wrap!

Callan Harrington 26:04
Thanks for tuning in! If you enjoyed the show, leave us a review—preferably on Yelp. (Kidding. But not really.)

Sullivan Finlay 26:19
We’re trying to get into Top 10 Best Corporate Comedy Shows in Chicago… on Yelp.

Callan Harrington 26:25
It’s a niche category, but we will dominate it.

Sullivan Finlay 26:37
Bye, everyone!