#199: AI Answers - Do Custom GPTs Still Matter? AI Output Validation, 2026 Job Disruption, Preventing Burnout, and Build vs. Buy (1h 1m)
ai-bias-fairness ai-driven-innovation-economy ai-in-everyday-life ai-in-workforce-disruption ai-literacy-public-awareness post-work-ai-society
- Release date: 2026-02-26
- Listen on Spotify: Open episode
- Episode description:
There is no shortcut for AI verification, and that's a good thing. Paul Roetzer and Cathy McPhillips answer 15 questions business leaders continue asking again and again. They unpack why AI output verification has no shortcut, where agent-building tools like Claude Code and Lovable actually stand, and the uncomfortable math behind which roles get disrupted next. Paul explains why enterprises are moving painfully slow even as the technology races ahead, how early adopters are creating burnout by doing the work of entire teams, and why situational awareness is the AI superpower most leaders are missing. 00:00:00 — Intro 00:07:00 — Question #1: Do you need to prompt AI the same way every time? 00:10:59 — Question #2: What problem do custom GPTs actually solve? 00:14:26 — Question #3: Are SaaS providers becoming model agnostic? 00:17:09 — Question #4: Why AI voice and tone change when models update. 00:20:36 — Question #5: AI output validation: why there's no shortcut for verification. 00:23:17 — Question #6: Tools for building AI agents: where to start. 00:26:11 — Question #7: Will knowledge workers face the same AI disruption as developers? 00:29:53 — Question #8: AI burnout: how leaders can prevent it during the AI transition. 00:36:21 — Question #9: Which roles and skills are most at risk from AI? 00:42:03 — Question #10: Traditional BI platforms vs. AI-first reporting systems. 00:45:22 — Question #11: Build vs. buy: AI decision framework for business leaders. 00:48:52 — Question #12: Competitive advantage for AI-forward agencies. 00:52:43 — Question #13: How to tell when someone just copy-pasted from ChatGPT. 00:54:39 — Question #14: Ads in AI platforms: what business users should know. 00:56:42 — Question #15: The one AI superpower every business leader needs. Show Notes: Access the show notes and show links here This episode is brought to you by Google Cloud: Google Cloud is the new way to the cloud, providing AI, infrastructure, developer, data, security, and collaboration tools built for today and tomorrow. Google Cloud offers a powerful, fully integrated and optimized AI stack with its own planet-scale infrastructure, custom-built chips, generative AI models and development platform, as well as AI-powered applications, to help organizations transform. Customers in more than 200 countries and territories turn to Google Cloud as their trusted technology partner. Learn more about Google Cloud here: https://cloud.google.com/ Visit our website Receive our weekly newsletter Join our community: Slack Community LinkedIn Twitter Instagram Facebook YouTube Looking for content and resources? Register for a free webinar Come to our next Marketing AI Conference Enroll in our AI Academy
Summary
- 🧠 Experiment Freely with Prompts: Balance structured prompts with simple ones across models for creativity; custom GPTs lock in consistency for repetitive tasks.
- ✅ Human-Led Verification: No AI shortcuts—humans must critically validate outputs to combat hallucinations and ‘AI slop’ flooding content.
- ⚙️ Enterprise Adoption Lag: AI disrupts coding now, but slow enterprises delay knowledge work transformation despite technical readiness.
- 🔥 Burnout in AI Era: Productivity surges overload early adopters; leaders need value-based rest and change management.
- 👁️ Awareness Superpower: Leaders lacking AI situational awareness hinder progress; deep understanding drives urgent scaling.
Insights
Should you always use structured prompting, or can AI carry context forward?
Time: 7:00 – 9:32
Category: AI Literacy & Public AwarenessAnswer: Structured prompting isn’t always necessary every interaction; custom GPTs or projects maintain instructions for consistency, but experimenting with simple vs. detailed prompts across models like Gemini, ChatGPT, and Claude often yields creative surprises as models evolve. This iterative approach fosters better AI collaboration without over-constraining creativity. (Start at 7:00)
What unique problems do custom GPTs solve beyond basic ChatGPT queries?
Time: 10:59 – 14:26
Category: AI in Everyday LifeAnswer: Custom GPTs ensure consistent output for repetitive tasks, pre-load context like instructional principles, and enable easy sharing for team use or public tools like JobsGPT, which has handled 30,000+ conversations. They productize prompts into accessible, specialized assistants for strategy, assessment, or brainstorming. (Start at 10:59)
Are SaaS providers shifting to model-agnostic approaches with bring-your-own-model?
Time: 14:26 – 17:08
Category: AI-Driven Innovation EconomyAnswer: SaaS companies like HubSpot don’t build models but access multiple providers (OpenAI, Anthropic, Google) as utilities, optimizing for cost by routing tasks to cheaper models while charging markups. Diversification enhances reliability and margins as model competition intensifies. (Start at 14:26)
Why do new AI models disrupt custom GPTs’ voice and tone, and how to fix it?
Time: 17:09 – 20:34
Category: AI in Workforce DisruptionAnswer: Model updates like GPT-5.2 prioritize agentic capabilities over writing, altering tone and performance; creators must re-test and tweak system instructions to restore desired personality. Users attached to prior tones face backlash, making ongoing maintenance essential. (Start at 17:09)
What’s the safest way to validate AI outputs and avoid ‘AI slop’?
Time: 20:36 – 23:16
Category: AI Bias & FairnessAnswer: Humans must remain responsible, verifying outputs critically without shortcuts like LLM checks alone; use AI for initial citation scrutiny but human sign-off prevents errors, hallucinations, and junk content flooding the internet. Slowing down ensures quality over quantity. (Start at 20:36)
Where should beginners start building AI agents like sales outreach tools?
Time: 23:17 – 26:09
Category: AI-Driven Innovation EconomyAnswer: Leverage existing tech stacks (Copilot, Gemini AI Studio, Salesforce) for rules-based agents before advanced tools like Claude Code or Lovable; agents excel in coding but remain rudimentary elsewhere, with rapid innovation making 2026 a breakout year. (Start at 23:17)
Will knowledge workers face developer-like disruptions from AI tools soon?
Time: 26:11 – 29:49
Category: AI in Workforce DisruptionAnswer: AI can theoretically automate most knowledge work now, but enterprise friction delays massive impact for 5+ years despite stock crashes in affected sectors. Urgency is needed as jobs decline before companies prepare. (Start at 26:11)
How can leaders prevent burnout amid AI-driven workload surges?
Time: 29:51 – 36:20
Category: Post-Work AI SocietyAnswer: AI supercharges output, turning early adopters into super-performers handling peers’ work, but requires change management: value-based rest (e.g., time off after high-impact projects), redefining productivity, and grace periods. Organizational design must adapt to intrinsic motivation. (Start at 29:51)
What skills will safeguard roles from AI obsolescence?
Time: 36:21 – 42:03
Category: AI in Workforce DisruptionAnswer: Emotional intelligence, interpersonal skills, problem-solving (asking right questions), critical thinking, and writing remain vital; no role is truly safe, as VC-funded AI targets lucrative labor markets, reducing human needs without eliminating them. (Start at 36:21)
What AI superpower would transform business leaders overnight?
Time: 56:42 – 58:45
Category: AI Literacy & Public AwarenessAnswer: Deep situational awareness of AI’s current capabilities and trajectory would spur urgent action: tool provision, training, centers of excellence, despite data fears. Most executives underestimate progress, stalling adoption 3+ years post-ChatGPT. (Start at 56:42)