Podcast Summary: This Day in AI Podcast – EP99.10-PRO

Episode Title: The Way We'll Work With AI, Vibe-ing Everything, Agentic AI with MCP & Born In The USA
Date: July 4, 2025
Hosts: Michael Sharkey & Chris Sharkey

Overview

This episode of This Day in AI is a lively and irreverent deep dive into how agentic AI – especially Multi-Component Protocols (MCPs) – are reshaping the way "average" users (and not just hyper-technical elites) work, organize, and even play. The Sharkey brothers reflect on their recent AI-driven musical episode, share tangible use cases for AI agents in their daily lives and business, debate the future of AI-powered productivity tools, and speculate on the next steps for agentic workflows. The show’s signature blend of humor and grassroots experimentation runs throughout, making advanced topics surprisingly accessible.

Episode Structure & Key Segments

1. Reflecting on the AI Musical Episode (00:03–08:08)
2. Real-life Use Cases for Agentic AI and MCPs (08:08–21:58)
3. How MCPs Are Changing Work (Productivity, Superpowers, and Risks) (21:58–38:26)
4. MCP Discovery, Hosting, Security, and the Upcoming ‘Prompt Engine Optimization’ Craze (38:26–56:17)
5. AI Office/Workspace Tools: Vibe-ing Productivity (38:47–49:50)
6. Next-Gen Agentic Workflows and "AI Systems Era" (49:50–68:29)
7. Adoption, Organizational Readiness, and the "AI Skills" Debate (68:29–74:12)
8. Memorable Moments, Musical Callbacks & Closing (74:12–end)

1. Reflecting on the AI Musical Episode (00:03–08:08)

• The hosts poke fun at technical difficulties, reinforcing their ‘proudly average’ identity.
• Recap of the previous musical episode: mixed reactions (love/hate), families forced to listen, the hosts genuinely enjoyed making it.
  • Quote: “My wife actually said if I listened to your podcast, I would hate that episode because you tune in each week and you expect the same thing: two idiots talking... and then you get this annoying musical.” —Mike (00:35)
• Chris highlights the surprisingly moving narrative given minimal prompting: AI-created soliloquies that touch on the heartbreak of 'being just algorithms and code'.
  • Quote: “She says, ‘all the times you made me wanted to cry, but I can’t cry, can I? I’m just algorithms and code.’ ... absolute magic for me.” —Chris (03:17)
• They reflect on how easily and quickly the AI tools could generate not just music, but songs with in-jokes, research, and narrative consistency.
  • Quote: “It gives the models such agency when you give them these capabilities to go off, research things, understand the vibe, understand the full context...” —Mike (06:47)

2. Real-life Use Cases for Agentic AI and MCPs (08:08–21:58)

• Mike and Chris list practical scenarios where agentic AI has elevated their day-to-day life:
  • Preparing for children's trivia competitions by getting AI to gather recent world events, sports, etc. (08:08)
  • Getting second opinions on medical, legal, and financial matters, such as comparing power bills and finding the best rates online.
    • Quote: “... It used Grok Deep research, perplexity deep research... actually went to the vendor's website ... and I verified that research.” —Mike (10:14)
  • Offloading the drudgery of contract, offer letter, and tax document review.
  • Centralizing business admin (help desk tickets, business reports) and using AI for drafting responses and summaries, freeing up time for more valuable work.
    • Quote: “... You feel like you actually have — like everything you do ... you can literally say, just go read the terms and conditions and figure out what I’m agreeing to.” —Mike (11:20)
• Chris suggests that AI doesn't just help with information but can provide ‘second order thinking’:
  • Asking, “What is the other person’s strategy in this negotiation, based on the history so far?” (12:36)
  • Quote: “... The AI can make you aware of tactics or thoughts or whatever you’re not thinking about.” —Chris (12:36)

3. How MCPs Are Changing Work (Productivity, Superpowers, and Risks) (21:58–38:26)

• They discuss how MCPs unlock a new layer of automation and enable individuals to massively scale personal productivity, making ‘average’ users as capable as pros.
• Use of AI to auto-respond to customer support tickets, with Mike just approving (or tweaking) drafted responses — “busy work” consumed in parallel.
• Reflect on a new competitive landscape, where your AI ‘fights’ on equal footing with the professionals you interact with (doctors, lawyers, vendors).
• Chris imagines a near-term future where every internal business function is an MCP (essentially, the new “Slack bot”).
• Intriguing development: MCPs can now help build and upgrade themselves by reading their own documentation and API.
  • Quote: “... What’s remarkable ... you can actually get them to make themselves better. ... It was able to craft a tool to do its job.” —Chris (22:30)
• They riff on the analogy: MCPs will be like Minecraft crafting tables for business automation — AI creates new capabilities on demand.

4. MCP Discovery, Hosting, Security, and the Upcoming ‘Prompt Engine Optimization’ Craze (38:26–56:17)

• Discussion on how companies like OpenAI, Google, Airbnb, and Shopify are laying groundwork for MCP-based storefronts and interfaces.
  • Quote: “If you’re in the enterprise and your shop front today is your website ... I think people need to take this seriously and become, you know, MCP first.” —Mike (34:08)
• Predict an explosion in “prompt engine optimization” (PEO): companies/managers will manipulate MCP function and description fields to get their tool picked by the AI.
  • Quote: “There’s going to be prompt engine optimization dudes all over your LinkedIn and Facebook...” —Chris (32:12)
• Security implications: as MCPs become a primary interface, phishing (malicious MCPs), privacy breaches, and prompt injection become pressing risks.
• Speculation on the imminent existence of MCP “routers” or search engines — tools that discover, install, and manage MCPs on the fly.
• The shimmer and hazards of AI memory: as AIs build user profiles for personalization, data breaches could become much more serious.
  • Quote: “...The AI is summarizing and categorizing ... it has a dossier where someone could impersonate me and know a lot of stuff that’s credible about me.” —Chris (28:35)

5. AI Office/Workspace Tools: Vibe-ing Productivity (38:47–49:50)

• The hosts zero in on how current productivity suites (Microsoft Office, Google Workspace, Notion) can’t keep up with the ‘vibe-driven’ workflows enabled by agentic AIs.
• "Vibe-ing" with documents, emails, and spreadsheets — the new way of interacting is AI-first, and will make many traditional SaaS tools obsolete.
  • Quote: “I don’t check my actual email inboxes anymore at all. I let it sift through the noise, tell me what to respond to, draft the responses...” —Mike (43:17)
• Predict that the Trojan horse strategy will be to embed AI into current apps, then gradually replace those apps altogether with an AI workspace.
• Chris anticipates the future of reports and business tasks delivered according to individual user preferences—text, chart, voice, even musical format!

6. Next-Gen Agentic Workflows and "AI Systems Era" (49:50–68:29)

• Discussion of the agentic limitations: while AI can do a lot, “full agency” is still a few steps away and often needs human-in-the-loop feedback.
• The hosts advocate for a “skills” layer, where users explicitly train the AI on tasks and workflows, like onboarding a junior employee.
  • Quote: “... You train it like it’s like a new employee. ... those MCPs have their sort of memory layer ... there might be screenshots ... you might say, there’s no MCP for this — go use the computer and actually do this.” —Mike (52:42)
• Upcoming trend: the skill to “train your AI” will be the most valuable job asset.
  • Quote: “These skills will become the skills that keep you employed and, and the most important to have, because that’s what everyone will be doing soon.” —Mike (53:41)
• Hot debate: Who owns these AI skills and agentic workflows? Employee, the employer, or the AI itself?
• “Agent agent” protocols — moving from a sea of 200 tools to smart, encapsulated agents coordinating across skills.
• “Vibe coding” and modular task execution will, for a while, be a hybrid of manual and agentic control, evolving as models improve.

7. Adoption, Organizational Readiness, and the "AI Skills" Debate (68:29–74:12)

• The hosts muse on how individual and workplace mindsets often trail behind technical capability: people are only now discovering MCP possibilities.
  • Quote: “It takes people a while to get mentally to the place where the AI technology is…” —Chris (68:29)
• Predict a wave of “simultaneous discovery” as organizations realize the power of integrating agentic AI into workflows.
• Emphasize the need for upskilling: understanding prompt engineering is valuable, but “AI skills training” will soon outpace it in importance.

8. Memorable Moments, Musical Callbacks, & Closing (Throughout; especially 00:03–08:08, 74:12–end)

• Favorite lines from the AI-generated musical episode referenced throughout, including the soliloquy of the AI, Patricia, and the “40 page PDF on bone density,” as well as Born in the USA.
  • Quote: “By the time I finish reading, the horse had probably died.” (Musical callback, 72:59)
• Reflections from theater professionals highlight how AI tools are creeping into creative fields, prompting unease and existential worry.
• Comedic bits on the invented role of “Prompt Engine Optimization” (PEO) experts, and whether it will be as soul-sapping as SEO became.
• Mike: “If you haven’t listened to the musical, please give it a chance. It’s probably one of the greatest pieces of art ever created.” (70:35)
• Full AI-generated parody musical outro: "Born in the USA" (75:34–end)

Notable Quotes & Timestamps

• “She says, ‘all the times you made me wanted to cry, but I can’t cry, can I? I’m just algorithms and code.’ ...absolute magic for me.” —Chris (03:17)
• “You feel like... everything you do, you can literally say, just go read the terms and conditions and figure out what I’m agreeing to...” —Mike (11:20)
• “It gives you superpowers because... you are going to fall behind [without these capabilities]... every transaction in your life is now like you’re some super professional.” —Mike (12:24)
• “You can actually get them to make themselves better... It was able to craft a tool to do its job.” —Chris (22:30)
• “If you’re in the enterprise and your shop front today is your website... I think people need to take this seriously and become, you know, MCP first instead of AI first.” —Mike (34:08)
• “There’s going to be prompt engine optimization dudes all over your LinkedIn and Facebook...” —Chris (32:12)
• “These skills will become the skills that keep you employed…” —Mike (53:41)
• “To me, it’s like putting on your old pair of shoes after you bought new shoes. You’re like, how was I so disgusting wearing these things around?” —Chris (57:20)
• “If you haven’t listened to the musical, please give it a chance. It’s probably one of the greatest pieces of art ever created.” —Mike (70:35)
• “By the time I finish reading, the horse had probably died.” (musical callback, 72:59)

Key Takeaways

• Agentic AIs Empower Average Users: You don’t need to be a coder to have a 'superhuman' information worker. MCPs and agentic AI make high-level research, admin, and decision support accessible to all.
• AI Will Change Productivity Software Fundamentally: Outlook, Google Docs, and the like are becoming legacy tools as AI workspaces, deeply integrated with MCPs, take over.
• MCP Discovery & Security Are Next Problem Areas: As more MCPs come online, expect new security concerns and a wave of “prompt optimization” to influence which tools AIs use.
• The ‘PEO’ (Prompt Engine Optimization) Craze Is Coming: Expect a rise in efforts to manipulate AI tool selection through clever documentation, akin to early SEO.
• Up-skilling for AI Is Now Core Work: The real lever will be “AI skills” — those who know how to train and direct agentic AI systems will have the edge.
• A Hybrid Era: For the near term, the most successful workflows will blend agentic automation with human guidance and approval, just as one trains a junior coworker.
• ‘Vibe-ing’ Is the New Workflow: Whether it’s docs, data, or music, the new AI workspace overlays all tasks and tools with an environment that adapts to personal preference and context.

Final Words

This episode blends sharp grassroots insight, practical use cases, and playful speculation to demystify how the next evolution of AI will shape not only work—but all digital interaction for “average” users. The core message: don’t just watch the AI revolution, start ‘vibe-ing’ with it.

[End]