The Menu Gen Test: Is Your App Obsolete?
Our read
The 'Menu Gen Test' is the brutal reality check every AI builder needs: if a single multimodal prompt can do what your app does, you're building a dead product. It's painful to kill projects, but it's better than watching them die slowly while you waste time on 'Software 1.0 plumbing' that the next model release will eat for breakfast.
Key findings
App Integrators: Adding AI features to existing apps provides value and extends their lifespan.
AI-Native Builders: If an LLM can do it, your app is redundant; focus on what only AI can truly create.
Developers are actively seeking frameworks and diagnostics to assess the viability of their AI projects amidst rapid model advancements, trying to avoid building solutions that will quickly become redundant.
What happened
The 'Menu Gen Test' is gaining traction as a blunt diagnostic for AI builders: can your app's core function be replicated by a single multimodal prompt and tool calls? This question is forcing developers to confront the imminent obsolescence of many AI-wrapped applications as models become more capable, leading to painful project culls.
The fight
- App Integrators
Adding AI features to existing apps provides value and extends their lifespan.
- AI-Native Builders
If an LLM can do it, your app is redundant; focus on what only AI can truly create.
The brief
The 'Menu Gen Test' is gaining traction as a blunt diagnostic for AI builders: can your app's core function be replicated by a single multimodal prompt and tool calls? This question is forcing developers to confront the imminent obsolescence of many AI-wrapped applications as models become more capable, leading to painful project culls.
The fight. App Integrators say Adding AI features to existing apps provides value and extends their lifespan. AI-Native Builders say If an LLM can do it, your app is redundant; focus on what only AI can truly create.
Why now. Developers are actively seeking frameworks and diagnostics to assess the viability of their AI projects amidst rapid model advancements, trying to avoid building solutions that will quickly become redundant.
From the episode. Karpathy's 2026 Playbook: Build Agent-First or Perish (https://www.youtube.com/watch?v=rsaaVXg28-8)