ChatGPT “apps” have been getting a ton of hype since OpenAI opened submissions in December. The pitch is simple: this is the iPhone App Store moment for AI — build once, tap into hundreds of millions of users, and ride the distribution wave.

In this episode, Jess and Kyle unpack what ChatGPT apps actually are (and what they aren’t). They break down the difference between apps, plugins, and custom GPTs, why the Apple comparison falls apart fast, and what the underlying architecture (MCP servers + in-chat widgets) means for builders who care about customer ownership, data, and monetization.

We also cover the buzziest news story in a while: Moltbook and OpenClaw (formerly “Clawdbot”), the viral “agents social network” story.

What You’ll Learn

What a ChatGPT app is and how it differs from plugins and custom GPTs
Why “App Store moment” is an oversimplification and what the real opportunity is
The mall kiosk vs storefront analogy: distribution without owning the customer relationship
Where ChatGPT apps genuinely reduce friction (and where they add it)
The practical constraints developers are hitting right now
How MCP changes the game for interoperability
What the Moltbook/OpenClaw incident reveals about security, hype, and “agent culture” narratives

TIMESTAMPS:

00:00 Introduction and Weekly Updates

06:41 ChatGPT App Store Launch and Overview

19:25 Understanding ChatGPT Apps vs. Plugins and Custom GPTs

28:57 The Model Context Protocol and Its Implications

33:00 The Future of AI Models and Ecosystems

36:05 Invisible Apps and Personal AI Agents

38:54 Navigating the ChatGPT App Submission Process

39:49 Exploring ChatGPT Apps for Users

43:02 Building ChatGPT Apps: Key Considerations

51:04 Evaluating the Viability of ChatGPT Apps

52:53 Moltbook and ClawdBot/Openclaw

Understanding ChatGPT Apps: Where They Help, Where They Don’t, and Why