Insights & Ideas

GPT-6 Astra crossed the AGI line and it changed almost nothing about your AI strategy

Individual users report 80% productivity gains from AI. Only 6% of organizations can tie it to earnings. GPT-6 Astra's benchmark score swings 37 points on the same weights depending on the infrastructure around it. The next model leap will not close that gap.

A company brain determines the return on every other investment you make

DBS Bank made a knowledge base the precondition for scaling AI and reported S$750 million in value while running an order of magnitude more models in production than peer banks. The difference was never the model. It was the company brain underneath it.

RAG as a service is the rational default for enterprise AI

Enterprises spent $37 billion on generative AI in 2025, but only 17% see real EBIT impact. The variable that decides outcomes is retrieval infrastructure, not the model, and it costs $40,000 to $80,000 to build before the second team to maintain it.

Your YouTube catalog is a knowledge base your audience cannot search

A channel with 10,000+ subscribers is a knowledge base locked in the one format that cannot be searched. After YouTube cut long-form recommendations 80%, the archive only compounds when audiences can query it directly.

RAG failure points are an architecture problem, not a model problem

RAG sits under 51% of enterprise AI, yet 80% of projects are projected to fail. A NAACL 2025 study found chunking configuration influences retrieval quality as much as the embedding model. The controlling variable is your data pipeline, not your model.

Your enterprise chatbot is only as good as the knowledge behind it

Vodafone moved first-time resolution from 15% to 60%. The enterprise chatbot knowledge base, not the model, decides whether your AI understands your company.