AI Explored Podcast Summary

Episode: Building a Team of AI Employees That Easily Scale Your Business
Host: Michael Stelzner (Social Media Examiner)
Guest: Gemma Bonham Carter (Founder, AI All Stars; Host, Office Hours with Gemma)
Release Date: December 30, 2025

Overview

In this episode, Michael Stelzner sits down with AI business expert Gemma Bonham Carter to unpack how business owners and marketers can create and scale teams of “AI employees”—not just using AI tools, but designing custom assistants that become reliable, productivity-boosting, business-specific team members. Gemma shares her journey into AI, major misconceptions, proven workflows, and a step-by-step framework for building AI assistants that drive business growth. Listeners get actionable advice on building robust knowledge bases, system instructions, and making AI work “on your team,” plus a peek at powerful new workflow tools.

Key Discussion Points & Insights

1. Gemma’s Path into AI and Why It Matters

• Gemma’s Background:
  • Started as a blogger in 2010, pivoted to online courses and digital marketing in 2016.
  • First big AI "aha moment" was using ChatGPT in 2022 to generate an Instagram caption:

    “In the moment, I thought, oh my gosh, like, this is wild that it is writing this so well in seconds. And I felt this whole kind of moment of everything is about to change.” — Gemma (03:11)

  • Hosted her first AI-focused summit in 2023 with 5,000 attendees, now 20,000+.
  • Her business now centers fully on AI-driven strategies, focusing on helping business owners “build out tools and AI-powered implementation resources.” (05:40)

2. Debunking AI Misconceptions

• Biggest Myth: AI assistants work well out of the box.

  “A generic assistant gives you generic output. So when you teach the assistant in depth about your brand, your audience, your frameworks, it then becomes one of the most reliable team members you can have.”—Gemma (06:43)

• Key Message: The setup is critical—“they’re blank slates” by default.

3. Tangible Benefits of AI Assistants

• Saving Time & Money:

  “My team... it’s me and two others who work part time, but we have been able to gain back, you know, 5, 10, 20 hours a week.” —Gemma (07:49)

• Creative Expansion:
  • No more blank-page anxiety; AI provides strong drafts, allowing for better ideation and consistency.
• Operational Consistency:
  • AI never forgets brand guidelines, ensuring stability and precision.
• Revenue Growth:
  • "We’re having our highest revenue year to date ever in my business. ... You can really leverage these for more profit...” (09:08)

4. Real-World Use Cases for AI Employees (12:00–15:30)

• Email Marketing Assistant:
  • Captures ideas on the go via voice notes and drafts full, on-brand email newsletters.
  • AI-generated emails are now better engaged, written faster, and preserve her unique voice.

    “I’m putting out more emails in less time. That gets better engagement than ever before.” —Gemma (12:13)

• Course Development Assistant:
  • Translates IP and lesson ideas into frameworks, worksheets, and course content swiftly.
• Others in Gemma’s AI Dream Team:
  • Podcast prep and production assistants
  • Customer support trained on specific programs
  • Platform-specific social media assistants
• Key Rule: If a task is repeated, you can (and should) build a specialized assistant for it.

5. The Brand Book: Essential Foundation (16:00–22:00)

• Step One: Before touching any AI tool, create a “brand book for your AI.”
• Contents:
  • Deep dive on target audience: “Trigger events, social/emotional desires, dream and dreaded scenarios, roadblocks, perceived risks, desired wins.”
  • Business backstory, personality, signature phrases/quirks, writing style.
  • Overview of offers, products, services, and company principles.

  “Most people skip this step... and they wonder why AI doesn’t sound like them or they’re not getting great results. ... You haven’t developed the depth of Brand Book that is really going to calibrate your assistants.” —Gemma (18:23)

• Format & Size:
  • Google Doc, exported as .md or .txt (not PDF).
  • Typically 50–100 pages (!), mostly generated via guided AI prompts (19:09).
  • Pie chart breakdown: 30–40% target audience, 40% owner deep dive/writing style, 10–20% offers (20:35).
• Advanced Pro Tip:
  • Consider adding customer reviews/testimonials for AI to reference (24:30).

6. Knowledge Files & Training Your AI Employees (25:35–32:15)

• Two Key Elements for Each Assistant:
  • Knowledge Files: Give deep context—writing samples, frameworks, case studies, SOPs, best practices (“playbooks”).
  • Instructions: Define the AI’s job description and behavior.
• Customization:
  • Brand book is shared, but knowledge files are custom per assistant (27:50).
• Building Expertise:
  • Use Perplexity, Deep Research (OpenAI), or Gemini for research to develop skill-specific knowledge bases, even if you aren't an expert (28:56).
  • Run prompts on multiple platforms and synthesize the best results (30:35).

7. System Instructions: Giving AI Specific "Job Descriptions" (32:16–37:06)

• Structured Under Four Headings:
  • Role: e.g. “You are a sales page copywriter for seven-figure brands.”
  • Context: Output specs, brand context, file references.
  • Behaviors: “How to act—if/then logic, preferred processes.”
  • Important: “Catch all” for reminders (“do not use EM dashes,” etc.)

  “With strong system instructions, this assistant then knows how to proceed, becomes reliable, very aligned with what you’re looking for.”—Gemma (34:53)

• Principle: One AI per trade: Specialize rather than generalize (e.g. a copywriter per platform/product, not one for all copy).

8. Daily Use, Collaboration, and Advanced Workflows (37:06–42:43)

• Integrating AIs into Habit:
  • Bookmark individual assistants for easy access.

    “Such a simple thing, but it means I can very quickly go to whichever assistant I’m looking for.” (37:57)

• Collaborative Workflow Automation:
  • Tools: Zapier, Make.com, Mindpal (for chaining/triggering assistants).
  • Google Workspace Studio now lets you create free, multi-gem workflows if you’re a Google Workspace user (40:20).
• Claude Skills:
  • Create “skills” (i.e., modular knowledge/instruction sets) for Claude, then invoke them in any chat (41:00).

9. Final Takeaways

• The True Power of AI Employees is in their deep, custom training—making them business-specific, creative, and consistent contributors.
• Documentation and Upkeep—consistent updates ensure AI assistants improve and evolve alongside your business.
• Phase Two: When you’re comfortable, explore workflow tools and multi-AI workflows to further automate and scale.

Notable Quotes & Memorable Moments

• On AI Transformation:

  “It wasn’t just like a new fun piece of software. It felt like this was going to be an entirely new way of operating.” —Gemma (03:25)

• On Building Assistants:

  “A generic assistant gives you generic output…they’re blank slates. So you need to invest the time into building out the back end.” —Gemma (06:43)

• On System Instructions:

  “We want AI assistants to be fairly specific in their role. ... I would rather have a copywriter trained on landing pages versus email marketing versus LinkedIn.” —Gemma (34:53)

• On Knowledge Base Building:

  “If you have this massive brand book and a lot of knowledge base information…write your instructions last...it’ll ensure that they’re aligned.” —Gemma (36:38)

• Habits for AI-Enhanced Work:

  “Building a habit of just going to them for every task you, your team is doing. ... Think about, could AI be helping with part of this here?” —Gemma (37:23)

Timestamps of Key Segments

• Gemma’s origin story & first AI summit: 02:26–05:40
• Debunking AI assistant myths: 06:16–07:22
• Benefits of AI assistants: 07:45–09:38
• Email Assistant and Course Development Use Cases: 10:48–13:03
• Gemma’s AI Dream Team examples: 14:15–15:30
• How to begin: Brand Book creation: 16:00–22:00
• Brand Book contents & formatting: 16:46–19:56
• Building deep knowledge bases: 25:35–32:15
• Structuring system instructions: 32:16–37:06
• Daily use and automation tools: 37:06–42:43
• Claude Skills and cross-assistant collaboration: 41:00–42:43

Resources & Links

• Guest Resources:
  • Instagram: @GemmaBonhamCarter
  • Free templates & brand book sample: gemmabottomcutter.com/SME
• Podcast: [Office Hours with Gemma] (show formerly “The Course Creator Show”)
• Host Resources:
  • AI Explored Show Notes
  • AI Business Society
  • AI Business World 2026 Conference

Key Takeaways for the Listener

• Effective AI “employees” require intentional setup: deep brand documentation, role-specific knowledge files, and clear system instructions.
• Customization, specialization, and ongoing refinement are critical to quality and scalability.
• Don’t be afraid to experiment—leverage tools for research, synthesis, and workflow automation as you grow your AI dream team.

For more deep dives and practical guides on putting AI to work in marketing and business, follow Michael Stelzner on your favorite podcast platform and check the show notes at Socialmediaexaminer.com/aipod.