#223: AI Answers - AI Washing, Flatter Org Charts, Advice for Students, Agent Security & the AI Writing Gap (57 min)

ai-driven-innovation-economy

ai-governance-laws

ai-in-workforce-disruption

privacy-in-the-ai-era

ai-driven-innovation-economy ai-governance-laws ai-in-workforce-disruption privacy-in-the-ai-era

Release date: 2026-07-02
Listen on Spotify: Open episode

Episode description:

The questions people ask about AI have changed. A year ago, they wanted to know what ChatGPT was; now they're asking how to redesign workflows around agents and whether a model can be trained the way a person is. In this AI Answers episode, Paul Roetzer and Cathy McPhillips take on fifteen of them, straight from our Intro to AI classes. From choosing two or three models on a tight budget to building an AI virtual twin without sacrificing authenticity, it's a snapshot of where practitioners' thinking actually is right now. 00:00:00 — Intro 00:5:59 How do you balance bottom-up experimentation with CEO-level strategy? 00:9:49 How do you move from restricting AI to enabling it? 00:11:36 How do you pick two or three models on a budget? 00:14:30 How do you evaluate vendors amid AI washing? 00:17:13 Frontier models, small models, or edge AI? 00:20:52 What are the security risks of autonomous agents? 00:25:02 Do AI models really behave like people? 00:30:25 How do you prove AI value with only basic tools? 00:32:02 How do you build a 24/7 AI virtual twin? 00:37:17 How do you close the human vs. AI writing gap? 00:40:26 Which skills gain value as AI takes over workflows? 00:42:57 Automate, augment, or keep it human? 00:47:20 Why flatten management instead of upskilling it? 00:50:51 Who's responsible for AI's economic fallout? 00:53:05 What advice would you give a college student? Show Notes: Access the show notes and show links here This episode is brought to you by AI for Business Bootcamp by SmarterX — a single-day event in Columbus, Ohio on July 16th, built for professionals and leaders ready to accelerate AI adoption and value creation. The day moves from a state-of-AI keynote into two hands-on workshops on AI productivity and AI innovation, and you'll leave with an actionable plan for yourself and your team. AI Academy members and groups get discounted pricing. Use code POD100 for $100 off your ticket. Learn more at SmarterX.ai/events. 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

⚖️ Top-Down Strategy Gap: Most organizations still drive AI from the ground up; CEO-level governance and clear policies remain rare but essential.
🔒 Agent Trust Dilemma: Autonomous agents unlock value only when deeply integrated, yet verification of data isolation is currently impossible.
💰 Rising Intelligence Costs: Token consumption is exploding; forward-thinking firms are already evaluating smaller open-source models for routine tasks.
🧠 Taste & Judgment Win: Domain expertise plus advanced AI fluency will separate high performers as AI handles more workflows.
📉 Entry-Level Disruption: Traditional entry-level roles are vanishing; students must become the top AI user in their chosen field to stay employable.

Insights

How can organizations balance grassroots AI experimentation with strong CEO-driven strategy and governance to avoid chaos while fostering innovation?
- Time: 6:00 – 8:09
- Answer: The discussion highlights that bottom-up experimentation is common but often lacks C-suite oversight on policies, risk, and future-of-work vision. This gap is widening as agents gain autonomy. Effective AI councils and transparent policies are presented as key bridges between the two approaches.
With frontier models becoming more expensive and agentic, when should companies shift to smaller open-source or local models instead of always using the biggest labs?
- Time: 17:26 – 20:11
- Answer: The transcript notes rising token costs and the emergence of cheaper Chinese models, prompting advanced organizations to evaluate task-specific SLMs. Most firms are still maximizing major platforms, but the conversation signals this cost-vs-capability trade-off will grow rapidly.
What new security and trust challenges arise when autonomous AI agents are given access to sensitive corporate data and tools?
As AI displaces both entry-level and knowledge-work roles, which skills and mindsets will become most valuable for long-term career success?