AI - Breaking Prod

Breaking Prod

Sign in Subscribe

AI

A collection of 9 posts

Anthropic Wants to Be a Utility When It's Convenient

And a walled garden the rest of the time. Anthropic tells governments that AI is so powerful it needs safety regulation. Dario Amodei (Anthropic's CEO) writes essays about how it will reshape civilization. The company spends millions lobbying for AI regulation because, in their own words, the technology

Nothing Your Agent Reads Is Safe

31 companies already tried to poison your agent's memory. You probably didn't notice. In February 2026, Microsoft's security team published research on what they call AI Recommendation Poisoning. Over 60 days, they identified 50 distinct attempts from 31 companies across 14 industries to plant

MCP Servers Are the Wrong Abstraction

For Coding Agents Every agent platform wants you to write MCP servers now. Anthropic launched the Model Context Protocol in November 2024, and within a year the ecosystem exploded. FastMCP tracks over 1,800 servers as of March 2026, up from 425 in August 2025. OpenAI, Google DeepMind, Microsoft, and

Beads Changed How I Work With Coding Agents

A colleague dropped a link in our chat a few weeks ago. "Have you heard of Gas Town?" "Mad Max Gas Town?" "Yes and no." The link went to Steve Yegge's Gas Town manifesto. If you haven't read it, it&

The Bottleneck Moves Up the Stack

First It Was the Code. Then the Specs. Eventually It'll Be the Humans. In my last post I wrote about code becoming an intermediate artifact, how the source of truth in software development is shifting from code to natural language as AI tools get more capable. That post

Stop Reviewing AI Code. Just Not Yet.

There's a day coming when reading AI-generated code will be as pointless as reading Java bytecode. Today is not that day. You should stop reviewing AI-generated code. I believe that. At some point, going line by line through AI output will be as useful as opening a .class

AI Is Breaking Open Source

And the Maintainers Are Done Asking Nicely. Last week, an AI agent published a hit piece on a matplotlib maintainer because he rejected its pull request. Not a human using AI to write code. An autonomous agent, running on OpenClaw, that researched the maintainer's personal coding history and

Vibe Coding Works Until It Doesn't

The Data Says You're Slower. You Just Don't Believe It. I use AI coding tools every day. Claude Code has access to my Kubernetes cluster. I've given it SSH access to debug hardware driver issues on machines I've never touched before. I&

Spec-Driven Development Tools Are Solving the Wrong Problem

I have been experimenting with spec-driven development (SDD) tools lately, and I keep running into the same issue. These tools are built for a workflow that does not exist in most organizations. If you are not familiar with them, SDD tools are AI-powered systems that interview you about what you