Strategic AI Custom GPTs Mitigating Bias

In this edition of AI Insiders by AI Marketers Guild, the host introduces Earle Richards Jr., who shares his expertise on leveraging AI strategically.

Earle explains how he uses custom GPTs as digital team members for career development, prospecting, and personal growth while addressing the importance of mitigating bias and personalizing AI responses.

0:05 – Host: Welcome to the latest edition of AI Insiders from AI Marketers Guild. We’re getting used to our new name and branding. At a recent event, someone commented that they couldn’t believe we got shirts made so quickly. (They’ll appreciate that story!)

0:32 – Host: I mentioned that I’ve had these shirts for 10 days—they’re already on their next laundry cycle. I promise I won’t wear just one when you see me.

0:58 – Host: Today’s focus isn’t me—it’s on learning from Earle Richards Jr. Many thanks to Mark Goldberg for introducing us. I was impressed learning new insights after just one conversation with Earle—even though I’ve known many community members for 20 years, Earle made an impact immediately.

1:20 – Host: For our newcomers, welcome to the community. Remember, we host these community conversations every week. I told the marketing team these aren’t webinars; these discussions are the foundation for our relationships.

1:44 – Host: It’s great to see familiar faces such as Katherine Montgomery and Jack Myers join us. I’ve mentioned Katherine’s session before—we may need an update on that soon.

2:05 – Host: That said, let’s welcome our guest. Earle, please introduce yourself.

2:29 – Earle: Hi, everybody. My name is Earle Richards Jr. I’m a fan of technology. I was born and raised in the U.S. Virgin Islands and moved to New York after seeing Home Alone—I knew I had to be here.

2:48 – Earle: I’ve been in advertising since college, working on agency, adtech, and publisher sides—always in the realm of data and analytics. This goes back to when we used Urchin before Google Analytics.

3:08 – Host: (Laughing) I haven’t heard that name in a while.

3:26 – Earle: I’ve worked across different verticals, brands, and agencies so I know what works technology-wise. People say I’m just a data guy, but I use data to win arguments—I trained in law for that reason.

3:45 – Earle: Today, I’ll present my approach to AI. David and I connected through Mark Goldberg and discovered we’re all trying to figure out how to use AI. There isn’t one “right” way—if someone tells you otherwise, they might be selling you something.

4:05 – Earle: I’ll show you what’s been working for me. I welcome questions and different perspectives to help me identify any blind spots.

4:22 – Host: I appreciate that you reject the term “like-minded.” It’s about bringing different points of view—such as those shared here from a group of 40 different users of AI.

4:44 – Host: So, Earle, can you tell us how you’re using AI these days and share some of your workflows?

5:10 – Earle: Absolutely. I’ve been curious about AI since I was a kid—after I broke my first computer, I learned how to fix it. When ChatGPT came out, I was eager to experiment and see what it could do.

5:36 – Earle: I’ve used AI to land jobs, reconnect with people, expand my network, and even get clients. The key point is: It’s not about the tool (ChatGPT, Gemini, or Perplexity). It’s about hiring AI as your digital team member.

5:54 – Earle: Back when I was a bellboy in high school, we had a concierge who always said, “I got a guy for that.” Now I say, “I have AI for that.” I ask: What do I want to accomplish and how can I build or use AI to get it done?

6:13 – Host: That’s an interesting mindset—the more you personalize your approach, the more effective AI becomes.

6:36 – Earle: Exactly. Whether you’re looking for a job or new clients, ask yourself: How can AI help accomplish your goals? People range from anxious that AI will take their jobs to overly excited that it will replace them. I prefer to consider AI as an ally.

6:58 – Earle: If you view AI as an enemy, you’ll be fearful. Instead, treat it as an extension of your team to help you achieve specific outcomes.

7:21 – Host: And, speaking of outcomes, I noticed you mentioned bias. How do you address that?

7:46 – Earle: Great question. Bias exists both in the training of AI engines and in the users’ preferences. Whether it’s the bias reflected in data or the bias of the person using it, you must first recognize it. For example, I train my AI to give clear, factual answers because that’s my preferred style.

8:07 – Earle: Different people have different styles (some prefer colorful language). Just as I teach my team how to work with me, you need to “teach” your AI your preferences by establishing guidelines or even user manuals.

8:29 – Host: That makes sense. Could you walk us through your process of prospecting a new client using AI?

8:50 – Earle: Sure. When prospecting, you research the client, identify who you know at the company, and position yourself for success. AI can automate much of that research, pulling together information from various sources.

9:08 – Earle: Think of it this way: If you want a new career or new clients, hire AI—as you would an assistant or expert. I’ve built about 40 custom GPTs for different purposes. For example, if I want to learn a platform like Adobe Analytics or GA4, I create a custom GPT expert on it.

9:36 – Earle: Instead of endless Google searches, you get a condensed, digestible delivery of the information you need.

10:01 – Earle: Then, if you want to use AI for business development, you build a custom GPT with a personality geared for prospecting and sales.

10:21 – Earle: Think of the difference between using cookies and not. Custom GPTs remember your preferences, continually tailoring their output as you interact with them.

10:40 – Earle: For instance, my career coach AI started generic, then I personalized it by adding my resume and work history. Now it can answer detailed questions from recruiters by pulling from a spreadsheet of my career achievements.

11:01 – Host: So it acts like a second brain for you.

11:22 – Earle: Exactly. It helps answer everything from cover letters to analyzing job postings. I advocate for stepping back, understanding your goals, and using AI strategically—whether you need personalized support for career shifts or for getting in shape.

11:43 – Host: (Laughs) There was mention in the chat about using custom GPTs for things like coaching through personal challenges. How far can we push this use case?

12:01 – Earle: People have asked about having different custom GPTs interact with one another. For instance, I built several experts for clients using various adtech platforms—they “talk” to each other to provide comprehensive research.

12:26 – Earle: I like to compare it to my own experience growing up in the ‘80s as an elder millennial influenced by action heroes. You need to decide: do you join Team Terminator or Team RoboCop? I’m on Team RoboCop—using AI as an ally.

12:50 – Earle: When working on prospecting, decide what you need to accomplish and then build your custom GPT with the expertise required. I have one for Google Analytics, another for GTM, and so on.

13:10 – Host: What distinguishes your custom GPTs from generic ones provided by ChatGPT or Gemini?

13:31 – Earle: The difference is personalization. It’s akin to using cookies versus not. The more you interact with and train your custom GPT, the more it learns your preferences. I even have one that’s been trained not to use fluff or overused phrases.

13:51 – Host: Could you give an example from your personal interests?

14:17 – Earle: Sure. I love movies and TV. I created a custom GPT ‘Movie and TV Expert’ that pulls in information from IMDb, Rotten Tomatoes, Fandango, and more. Ask it, “How much money did Home Alone make?” and it provides detailed box office numbers, historical context, and even fun facts like Kevin’s paycheck.

14:38 – Host: Does a generic ChatGPT give you the same detailed response?

15:02 – Earle: It might give a generic answer, but my custom GPT is tailored to my enthusiasm and knowledge level—it gives you that director’s cut answer.

15:23 – Host (Lisa’s question): What’s the source for that detailed information? Do you feed it manually or does it pull from the internet automatically?

15:47 – Earle: I include specific sources and references—IMDb, Wikipedia, YouTube, Box Office Mojo, and others—in the setup for the custom GPT. I compile resources using files, PDFs, and online content so the system can pull accurate, verified details.

16:08 – Host: That explains a lot. How do you handle integrating additional resources like PDFs or wikis?

16:27 – Earle: When building a custom GPT, I upload any relevant resources—be it spreadsheets, PDFs, or reference sites. The system then uses this background to generate responses that are tailored to your specific context.

16:47 – Earle: For instance, when I built my career coach GPT, I added my entire work history, personality test results, and client experience details. This ensures that when a recruiter asks about my expertise, the answer is detailed and personalized.

17:08 – Host: It sounds like you’re using AI as a true second brain.

17:37 – Earle: Yes, exactly. I use it to consolidate information and generate insights for job applications or client proposals. It’s about leveraging AI to accomplish your goals faster and smarter.

18:02 – Earle: I must stress that as you teach your custom GPT, you’re effectively training a team member. Every piece of added detail helps personalize the output.

18:24 – Earle: I even feed it my personality traits (I’m direct, clear, and hate fluff) so that its answers reflect my style.

18:44 – Host: That’s a clear illustration of using AI to reinforce your brand and personal approach.

19:06 – Earle: Exactly. Think of it the way you’d hire someone—what qualities do you need? Then train your AI accordingly. It becomes an extension of you, whether for career guidance, research, or even personal interests.

19:26 – Earle: For example, I built a custom GPT called “G4 Expert” that I use to answer any questions related to Google Analytics 4. I fed it free online resources so that it can serve as a reliable studio assistant.

19:41 – Host: And this customization saves you time, right?

20:07 – Earle: Precisely. Instead of manually searching for answers across blogs and FAQs, my custom GPT aggregates and analyzes the data to give me a concise answer.

20:37 – Host: I see chat users are curious about practical examples in career consulting. Could you detail how you use custom GPTs for that?

20:59 – Earle: Sure. I once built one specifically to analyze job postings. I would copy the job description, and it would break down why the job fits my skills or show what might be missing. It acts as my career consultant—a second brain to review my resume and draft custom cover letters.

21:19 – Earle: I even prepared a spreadsheet listing my clients, platforms I’ve worked with, and years of experience on each. By integrating that data, my GPT can answer even technical questions from recruiters.

21:42 – Host: It’s like outsourcing part of your own knowledge. Do you ever worry about sharing personal data with these platforms?

22:04 – Earle: I work in data, so I understand the risks. It’s a balance between convenience and security. I personally lean toward sharing enough data to get a truly personalized experience while understanding that if someone really wants your data, they can get it.

22:28 – Host: That’s reassuring and honest advice. We all have our different comfort levels with data sharing.

22:50 – Earle: Exactly. It depends on your context—if your work centers on security, you need to be more cautious. But for me, the benefit of working faster outweighs the potential risk.

23:13 – Host: Let’s shift a bit. Some of our participants want to know if people are charging for these custom GPT services. What’s your take?

23:34 – Earle: Yes, I am—and I know others charging for custom GPTs. I always start by understanding a client’s pain points and goals regarding AI use. Custom GPTs can save you thousands on generic coaching by providing personalized, tailored advice.

23:57 – Host: That makes sense. Could you share any price range without revealing too much?

24:19 – Earle: I’m not there yet publicly, but I take inquiries and tailor services based on specific needs. We can discuss details offline if you’re interested.

24:42 – Host: And what about the interface? Does your client interact with a polished app or within the native OpenAI interface?

25:16 – Earle: It depends on the client’s needs. I default to the simplest interface available for ease of use. For high-security clients, integration within internal systems is possible. The important thing is to start where you are comfortable and build it up gradually.

25:31 – Earle: Remember, AI isn’t one-size-fits-all. You need to build, teach, and train it—just as you would coach a new employee.

25:44 – Host: Before we wrap up, I’d like to ask: How can people follow you and continue this conversation?

26:08 – Earle: The best way to reach me is through my Linktree (my contact info is in the description) and my LinkedIn profile. I’m happy to set up one-on-one conversations. Whether it’s for business or personal use, I’m here to help you apply AI in a way that makes sense for your goals.

26:33 – Host: That’s fantastic. I also love how you mentioned reverse mentoring—learning from each other even across generations.

26:55 – Earle: Absolutely. After speaking with some of you, I’ve come to fully embrace my “co-pilot” AI. It helps soften my tone and even offers fresh perspectives I hadn’t considered. I encourage everyone to connect with someone who has a different viewpoint.

27:16 – Host (Amy): I’ve been using AI as a thought partner for a while, and sometimes it challenges my own biases. It’s incredible.

27:38 – Earle: Precisely—AI always answers your question based on the prompt it’s given. That’s why building your own knowledge base is crucial. That way, you know what to expect and can verify the answers.

28:02 – Host: I love that concept of a “second brain.” It reminds me of Thiago Forte’s method of organizing ideas digitally.

28:26 – Earle: Yes, I borrowed that concept. My custom GPTs help me store frameworks, mental models, and insights from books and conversations—all of which gets synthesized into actionable advice.

28:49 – Host: And that is vital in our age of information abundance. Distinguishing the signal from the noise is key.

29:08 – Earle: Exactly. We live in an era where automation and data overflow require critical human skills—like creativity and synthesis—that AI can enhance but not replace.

29:34 – Host: I appreciate your clarity on merging human skills with technological advances.

30:05 – Earle: Thank you. It’s all about using AI to augment what we do best, while never losing our uniquely human touch.

30:27 – Host: Before we finish, any final thoughts on how our community should approach AI?

30:48 – Earle: Remember, it’s not about using a tool—it’s about building a team. Train your AI just as you would an intern, assistant, or a coach, so it truly represents you and supports your goals.

31:10 – Host: That’s truly inspiring. I’m sure many viewers will reflect on how they can personalize AI in their own lives.

31:32 – Earle: I appreciate all the feedback. Whether you’re using AI for career development, client prospecting, or even just exploring personal interests, let it work for you rather than against you.

31:50 – Host: Thank you all for your great questions and comments. I’ll compile the excellent book references and recommendations from the chat and share them with the community later.

32:07 – Earle: Thank you for this engaging conversation. Remember, if you want to learn how to use AI as a personalized thought partner, just reach out. I’m always happy to offer 30-minute sessions to help you get started.

32:35 – Host: And next week, we have Chris Huer coming in from a Teamflow perspective—don’t miss that. We also hope to see some of you in person during Tech Week.

33:12 – Earle: It’s been a fantastic discussion. Thanks to everyone for your questions, insights, and willingness to learn. Let’s keep the conversation going on LinkedIn and beyond.

33:36 – Host: Thank you, Earle, for inspiring us today. Goodbye everyone and see you next Wednesday at noon Eastern!