DiscoverChaos Lever PodcastWhy Your AI Assistant Still Sucks (And How MCP Might Help) | Chaos Lever
Why Your AI Assistant Still Sucks (And How MCP Might Help) | Chaos Lever

Why Your AI Assistant Still Sucks (And How MCP Might Help) | Chaos Lever

Update: 2025-04-10
Share

Description

This week’s main dish? Agentic AI and the Model Context Protocol (MCP). What the heck do those mean? Why are they being compared to USB-C? And why should you care unless you’re an executive with a robot butler? Ned breaks it all down while Chris offers the occasional therapy check-in. Spoiler alert: MCP is the plumbing behind smarter AI assistants, but whether we trust them with our calendar (or our lives) is still up for debate.

Oh, and yes, there’s a “Silver Spoons” reference, some Carlton love, and a side quest into RESTful APIs because this is Chaos Lever and we can’t stay on the rails. Literally. We try to unpack whether MCP could be the REST of the AI world or just another shiny-but-useless indoor train. Buckle up.

🔗 LINKS
Model Context Protocol: https://modelcontextprotocol.io/introduction
The Train: https://external-preview.redd.it/T4x6zmXqtoaJQxw8uhtcNdquSLFHualiTg1Gnac_ihA.jpg?auto=webp&s=6b728fb53bfab7cbb77d1bc54714f9362d33c4b5

Comments 
In Channel
loading
00:00
00:00
1.0x

0.5x

0.8x

1.0x

1.25x

1.5x

2.0x

3.0x

Sleep Timer

Off

End of Episode

5 Minutes

10 Minutes

15 Minutes

30 Minutes

45 Minutes

60 Minutes

120 Minutes

Why Your AI Assistant Still Sucks (And How MCP Might Help) | Chaos Lever

Why Your AI Assistant Still Sucks (And How MCP Might Help) | Chaos Lever

Ned Bellavance and Chris Hayner