I Made Opus and Fable Grade Each Other. Opus Admitted It Lost.
The creator demonstrates real-world AI applications in their productivity business using Claude Fable and Opus, showing how Fable excels at comprehensive analysis of complex systems while comparing its performance to Opus on specific tasks. The key finding is that Fable's holistic understanding justifies its higher cost for business-critical audits and optimizations.
Summary
The video documents the creator's practical use of Claude Fable 5 API access for auditing and optimizing the ICOR Journey courses—a comprehensive productivity education platform with interconnected lessons spanning note-taking, PKM, task management, and automation. The creator emphasizes real business applications rather than toy examples, explaining how their productivity system includes deep course content (up to 30,000 words per lesson), growth assignments, explainer videos, and cross-referenced connections.
The primary use case involves auditing the courses to ensure proper "golden thread" progression where earlier lessons are referenced by later ones without forward references that would confuse learners. Using Claude in VS Code with custom prompts, the creator generates interactive HTML visualizations using React Flow showing course dependencies and connections—a capability they demonstrate works across complex knowledge bases.
The critical comparison emerges when the creator tests both Fable and Opus on the same tasks. For a password reset issue diagnosis, Opus completed analysis in 3 minutes while Fable took 8 minutes, but Fable discovered a more elegant, proper solution rather than just a UI workaround. The creator demonstrates this by having each model review the other's session, with Opus acknowledging that Fable "went one layer deeper" and found superior solutions.
The creator addresses the cost-benefit analysis explicitly: while Fable API access is expensive (costing potentially thousands of dollars for comprehensive audits), the return on investment is clear when it prevents wrong conclusions in complex system analysis. They draw parallels to hiring human experts who also produce variable quality outputs and can become bottlenecks. The folder structure and standard operating procedures the team uses enable consistent results while allowing AI systems to think outside the box differently.
Throughout the video, the creator emphasizes that success with AI depends on having proper foundational systems (single source of truth, clear information architecture) in place first—attempting to automate broken processes only amplifies problems. The real power emerges when AI is layered on top of already-optimized workflows.
Key Insights
- Fable's advantage over Opus is not speed but depth—it took longer (8 minutes vs 3 minutes) to analyze a password reset issue but discovered a proper architectural solution rather than just a UI workaround, and Opus acknowledged that Fable 'went one layer deeper'
- Fable justifies its high API costs for business-critical analysis because it prevents wrong conclusions that could send entire teams down incorrect paths, similar to how human experts on billion-dollar projects can cause significant losses with faulty analysis
- The creator's return on investment calculation for expensive Fable audits depends entirely on whether optimizations will improve user retention and content completion rates, requiring understanding of overall business numbers rather than just token costs
- Fable excels at analyzing complex interconnected systems holistically—understanding how all 168 gigabytes and 160,000 files in a knowledge base connect together—which is where the creator sees the real competitive advantage over other models
- AI layered on top of already-broken processes only amplifies problems, so proper foundational systems with single source of truth and clear information architecture must exist first before automation or AI optimization can be effective
Topics
Transcript
[0:00] Fable 5 is back. I'm really happy about it. The only thing that I see on YouTube and other channels, whenever people talk about these models, I rarely see real use cases where they are - getting applied in real life work. And that's why in this video I will show you the first thing that I did in my business the moment I got access to Fable again and I can tell you already, Fable didn't disappoint. It is really insane, And I really hate the fact that we will only get access through API to Fable in the future because that's very expensive and that's why I [0:33] take this time where I have access to it…
Full transcript available for MurmurCast members
Sign Up to AccessMore from ICOR with Tom | AI Productivity
Our AI invented the numbers. We caught it.
The hosts discuss the importance of double-checking AI outputs and validating information, emphasizing that this is not a new problem but rather a longstanding practice required with any information source. They explore how AI tools can be customized to individual workflows and contexts, arguing that understanding foundational methodology matters more than specific tools.
An AI Just Handed Me a Fake $67B Statistic
The speaker shares how an AI confidently generated a completely fabricated $67.4 billion statistic while researching AI hallucinations, then explains his systematic approach to combating AI misinformation: using dual independent search engines to cross-check answers and identify where hallucinations occur.
AI Is Quietly Taking Your Memory
AI companies are racing to own users' context and memory through integrated tools like Claude in Slack and ChatGPT's auto-memory features, creating a lock-in trap. The speaker argues users should own their context (expensive, irreplaceable) in local plain text files while renting the AI model (cheap, replaceable), and demonstrates his personal knowledge assistant system built on markdown files that work with any AI model.
My AI Team Now Has an Interface. All 12 Agents. Free.
Tom, creator of the iCola methodology, releases My PKA version 3 with a free interface (My PKA Cockpit) that provides 12 pre-built AI agents for productivity, note-taking, health tracking, document management, and task planning—all running locally in a folder structure accessible via Claude and a web interface.
They Restricted Fable. I Switched Models in One Folder.
The creator explains why the removal of Claude Fable didn't disrupt his workflow, arguing that model-agnostic local folder setups with structured instructions make users resilient to AI changes. He uses an analogy of a dark room with a chair to explain how context and instructions guide AI behavior. The core message is that building robust local documentation and agent orchestration systems matters more than chasing the latest AI model.