Anthropic expands Cyber Verification Program into three access tiers

The Trump administration is using legal liability rather than regulation to enforce AI safety, leaving unresolved who pays when models go rogue: the companies that design the systems or the people who deploy them.

Talus generates 64x64 heightmaps (4 km, up to 1,200 m) conditioned on terrain type and any subset of five measured properties: mean elevation, relief, mean slope, water fraction and spectral slope. Trained from scratch on one RTX 5060 (8 GB) in about 4.5 hours total, and evaluated against a real-vs-real noise floor.

Community GGUF quantization of Qwen-Image-2.1 Turbo, an uncensored image model, drew 102 likes and 7,314 downloads on HuggingFace.

Standard Bots raised $200M at a $1B valuation in a Series C led by General Catalyst and RoboStrategy, with customers including NASA, Amazon and Lockheed Martin. Its largest model is in the low billions of parameters, and Head of AI Leif Jentoft says data quality matters more than raw volume.

Video discussion weighs the business models of AI labs against traditional SaaS companies heading into 2026. No specific figures or companies are detailed in the available source material.