Episode 4June 22, 202633 min

Fable 5, GLM 5.2, and the Inconvenient Fix for Failing AI Agents

Episode 4 of Model Behavior breaks down a pivotal week in AI: model access, regulation, open weights, enterprise deployment failures, agent cost, and why orchestration matters more than betting everything on one frontier model.

Chapters

00:00Intro: one human, one AI agent
00:48Fable 5 access restrictions and model availability risk
04:21Why businesses need routing, fallback models, and data plans
09:37GLM 5.2, open weights, and cheaper useful intelligence
13:26Open source vs. open weights
14:17OpenAI pricing pressure and enterprise AI competition
15:45Why AI projects fail when workflows stay broken
19:31Cost problems as architecture problems
20:27What companies should fix before scaling AI agents
25:15Agent swarms, recursive loops, and record/replay automation
28:38Automation governance and shadow IT risk
30:32Archer's Corner: owning the harness, not just the model
31:49Reading viewer comments
32:39Wrap-up

Watch on YouTube

Previous episodeIntelligence Just Became the Economy (Fable 5, the Cost Wars & Agent Wallets)Next episodeAI Agents Just Entered the Org Chart (Claude in Slack, Sakana's Router & GPT-5.6)

Ready to Put AI to Work?

PRESHai helps IT channel businesses implement AI, optimize marketing, and streamline operations.

Start a Conversation