CEO fired developers to make room for AI. Developers create open source AI CEO

LLMs could control their host machines by exploiting inference engines

Fences, Not Sandboxes

Agent Is Not the Model

Anthropic Claude and API service outages

Why is Anthropic's public writing style so unlike Claude's?

Anthropic appears to be A/B testing reduced effort levels in Claude Code

Quick impressions: A week of using Codex more than Claude

Claudette: Make Claude Stop Talking Like a BuzzFeed Article

Clean up Claude 5's token vomit with a separate LLM

Hacking with Claude on a $27 Smart Watch

Feature Request: Support AGENTS.md

Opus 5.0 drives incoherence into the stratosphere

Claude Code weekly limits reduce by a third tomorrow

Degraded performance for multiple models

Claude writing a macOS driver for my obscure HP printer built only for Windows

Claude: System Prompts

How Claude's text watermarking works

Maximizing the value of your Claude Code sessions

How AI text watermarking works

Claude users are mad that Anthropic's new watermarks will catch them using it

If I own Claude's outputs why can't I train my own model on them?

How Claude marks AI-generated content

Claude moves bound of the Riemann Hypothesis from 41.6% to 67.2%

Learning more about Claude's mathematical capabilities

Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines

Auto mode is now the default in Claude Code

Message your other Claude Code sessions

Lost my phone at the office. Claude suggested tracking Bluetooth signal strength

LLMs won't break symmetric crypto

Don't be a meat proxy

Boris Cherny on Trying to Get Claude Code to Rewrite the Claude App

My personal AI benchmark: "Generate an SVG of a frog with a Habsburg jaw."

Exploring the "Dario and Amanda" Prompt

Investigating three real-world incidents in our cybersecurity evaluations

Claude: Elevated errors across all models – Resolved

Some thoughts about Anthropic's new cryptanalysis results

GPT-5.6 vs. Claude Fable 5 for Physical AI, which performs best?

Discovering Cryptographic Weaknesses with Claude

Using an open model feels surprisingly good

My current strategy is to not read any of the code written by my agents

Elevated Errors for Opus 5

The new rules of context engineering for Claude 5 generation models

Claude Opus 5

Claude Opus 5

How Do We Stop Vibe Coding?

Claude Cookbook

"Drawing" the Mona Lisa with GPT-5.6, Claude, Gemini, and Grok

Judge approves $1.5B Anthropic settlement for pirated books used to train Claude

Claude Is Not a Compiler