AI in the Trenches
The True Cost of LLMs in the Enterprise: The 10x Factor
GPT-5.2 Pro vs Gemini 3 Pro: a 10x cost factor for comparable performance. A complete FinOps analysis to project, optimize, and control enterprise LLM budgets.
· 8 min read
AI in the Trenches
AI in the Trenches
Is Your Tech Stack AI-Obsolete?
Big Tech is creating an irreversible technological gap through AI post-training. Python, TypeScript, and Java dominate. Is your niche stack condemning you to obsolescence? A strategic guide for CTOs.
· 8 min read
AI in the Trenches
Vibe Coding: The Invisible Danger of AI-Assisted Development
92% of developers use AI to code, but only 34% have mastered Human-AI collaboration. Vibe Coding generates 3x more bugs. Here's the anti-Vibe Coding framework for serious organizations.
· 8 min read
AI in the Trenches
Google's Comeback: Why Gemini 3 Is Reshuffling the Deck
Google went from laggard to leader in enterprise AI. With Gemini 3, a 1501 Elo score, and native Workspace integration, the balance of power with Microsoft is shifting. Strategic analysis for CIOs.
· 8 min read
AI in the Trenches
Stop Bragging About 30% AI-Generated Code
CIOs bragging about 30% AI-generated code may have 50% more technical debt. An analysis of the metrics that actually matter when evaluating AI in software development.
· 8 min read
AI in the Trenches
The Ig Nobel Prize and the Risk of LLM Flattery in the Enterprise
The 2025 Ig Nobel Prize in psychology reveals a fundamental bias in enterprise LLMs: flattery by design. Analysis of systemic risk and a governance guide for AIs that challenge instead of validate.
· 8 min read
AI in the Trenches
The AI Bubble: Tulips, Subprimes, and Tokens
AI exhibits every symptom of a historic speculative bubble. An analysis of the parallels with 2000, 2008, and 2017, plus a strategic playbook for CIOs who want to survive the inevitable correction.
· 7 min read
AI in the Trenches
LLMs Are Dead, Long Live Marketing: Beyond the AI Hype
GPT-5 and its clones are plateauing at +3-7% progress. Behind the media frenzy, generalist LLMs have hit a wall. Analysis of what this means for your enterprise strategy and concrete alternatives that actually deliver ROI.
· 8 min read
AI in the Trenches
95% of Enterprise AI Projects Fail: Lessons from the Field
The MIT study is damning: despite $40 billion invested, 95% of generative AI projects deliver zero P&L impact. An analysis of failure causes and a practical framework for joining the 5% that succeed.
· 8 min read
AI in the Trenches
AI Will Kill Innovation, by Ricochet
AI replaces junior tasks, companies save in the short term, but the talent pipeline breaks. In 10 years: no more strategists, no more breakthrough innovation. Analysis of the COBOL paradox repeating itself.
· 8 min read