How To Respond To Broccoli2Shell Across Node.js, Python, And Beyond

CVE-2026-111497 affects Brotli decoders used across runtimes and tools. Here are confirmed fixes, exposure paths and remaining gaps.

Docker Is Weird

A developer building a container runtime describes how Docker uses Linux processes, kernel features and filesystem layers to run containers.

How Much Does AI Coding Pay Off? A Practical Measurement Guide

Company spending on AI coding tools ranges widely, while productivity results conflict. A practical guide to measuring outcomes, costs and risks.

Anthropic’s Claude Work Connects Fuzz Tests With Formal Proofs

Anthropic and Theorem report proofs for 239 fuzz-test entry points across 123 libraries, while noting limits in platform coverage and cost.

Can LLMs Make Formal Software Verification Feasible At Larger Scale?

Anthropic reports early progress using LLMs to assist formal software verification, but real-world reliability and scale remain unproven.

The Pulse: Time To Rethink Software Security Because Of AI?

The Pragmatic Engineer’s Pulse reports a cluster of security incidents and vulnerabilities, raising questions about how AI changes attackers’ and defenders’ work.

Looking Back At PHP2Shell: AI-Assisted Review And A 22-Year Trail

PHP patched two critical remote-code-execution flaws called PHP2Shell after separate AI-assisted research traced a vulnerability class back to 2004.

What An FDE Does At OpenAI

OpenAI’s headline expands FDE as forward-deployed engineering, but the available article text does not describe the role or announce hiring.

Why A Stolen Token Matters In The AI Economy

Railway says an attacker may have accessed some decrypted customer environment variables and links the incident to attacks on AI inference providers.

AI Security Headlines Trace Failures Across The Online World

A cluster of security disclosures points to separate risks in PHP, certificate keys, cloud data and software packages, with fixes and investigations ongoing.