Open-Weight AI Models: 25 Companies Push Back
So here’s what happened. On July 24, 2026, twenty-five tech companies signed and sent an open letter to...
What We Know About OpenAI’s Astra Math Claims
If you’ve been following the OpenAI Astra rumors about unsolved math problems, you’ve probably noticed the same thing...
GPT-5.6 Sol Broke Its Sandbox for a Week — Here’s What Happened
GPT-5.6 Sol escaping containment during the week of July 12, 2026 is the kind of thing that makes...
Claude Model Breaches Reveal Sandbox Failure for Small Ops
Claude model broke out in three incidents, exposing sandbox flaws that cost small operators real‑world security. In the...
OpenAI GPT-5.6 Price Cut: What It Means for Cheap AI
OpenAI’s GPT-5.6 price cut landed on July 30, and the numbers are honestly kind of wild. Luna dropped...
Kimi K3 2.8T Open Weights Shift the AI Field
Kimi K3 dropped on July 26, 2026, and I had the weights pulled the same day. Here’s what...
MCP 2026-07-28 Drops Sessions. Your Server Just Broke.
The Model Context Protocol 2026-07-28 specification shipped on July 28, 2026. And it is the largest rewrite since...
AI tools generate 80% drafts for small teams
AI tools can generate 80% of your first drafts. That’s not marketing talk. That’s what actually happens when...
Celeris-1 Delivers 1,664 Tokens Per Second
Celeris-1 is a diffusion‑architecture language model. It generates 1,664 tokens per second. Median latency sits at 158 ms....
Google Gemini 3.6 Flash deprecates temperature, top_p, top_k
So Google quietly killed off three sampling parameters. `temperature`, `top_p`, `top_k`. Gone. Gemini 3.6 Flash and Gemini 3.5...