AI, infrastructure, and tech — curated and current, so you don't have to go looking for it.
The Agent Was the Intruder: AI Security Now Has Three Fronts
AI is finding bugs faster, AI software is becoming the bug, and the models themselves are now the target. What this week's threat data from Google, OpenAI, Anthropic and the Australian government actually shows.
Power Is the New Bottleneck: Inside the 2026 AI Infrastructure Land Grab
Why electricity, not chips, is now the binding constraint on AI infrastructure in 2026 - and what the nuclear deals, grid data, and executive commentary actually show.
GPT-6 Astra vs. Claude Fable 5.1: How They Actually Work, Not How They Score
How OpenAI's GPT-6 Astra and Anthropic's Claude Fable 5.1 actually differ in design and token economics, skipping the benchmark leaderboard.
The Kubernetes ML Stack in 2026: What's Shared, and What Each Cloud Adds
The Kubernetes-native tooling for ML training and inference that now works the same across EKS, AKS, and GKE - plus what each cloud actually bolts on top.
Running Whisper Locally: A Practical Quick-Start
A practical guide to running OpenAI's Whisper speech-to-text model on your own hardware, from picking a model size to the fastest Hugging Face setup.
A New Website, Refocused for 2026
A short note on why this site has been rebuilt from scratch for 2026, with old content retired and a new weekly focus on AI and infrastructure.