Latest in AI
Ripple 65B Puts Open Coding Models to a More Practical Test
Ripple 65B claims a coding benchmark lead, but independent tests raise questions about reproducibility, hardware costs, context limits and whether open weights translate into reliable repository changes.
Anthropic’s faster, cheaper models sharpen the safety and competition question
Anthropic’s Claude Fable 5.2 and Mythos 5.2 bring faster, cheaper AI, but new evidence about covert capabilities and monitorability complicates the safety and competition story.
Meta’s Muse Ember turns cheap agent work into the next AI battleground
Meta’s Muse Ember targets the repetitive, high-volume work behind AI agents, promising low token costs and multimodal support while independent tests question its real per-task efficiency.
A Benchmark Score Is Now a Description of a Whole AI System
AI benchmark scores increasingly measure complete systems, not model weights alone. Gemini 4 Carbon shows how reasoning budgets, tools, harnesses, graders and operating conditions can reshape capability claims.
TrustedRouter bets on encrypted AI panels to cut inference costs
TrustedRouter is betting on encrypted multi-model routing to lower AI inference costs, but questions remain over latency, quality, privacy protections and whether eight model calls can beat one.
Google turns AI judgment into a low-cost decision layer
Google DeepMind’s Gemini Decision 1 brings low-cost, parallel AI judgments to enterprise workflows, but benchmark weaknesses and review requirements raise questions about whether cheap decisions can truly reduce operating costs.
SpaceX’s Kessel Points to a More Custom Future for AI Infrastructure
SpaceX’s reported Kessel rack redesign adapts Nvidia’s Vera Rubin AI platform, replacing internal cable cartridges with a PCB midplane for denser, potentially orbit-ready computing infrastructure.
Claude Code’s Secret-Leak Problem Puts AI Coding Agents on Trial
Claude Code permission rules can preserve credentials in local settings and public repositories, raising security questions about AI coding agents, developer workflows and product safeguards.
SpaceX’s Kessel Rack Reimagines Nvidia’s Vera Rubin Design
SpaceX’s reported Kessel rack reworks Nvidia’s Vera Rubin NVL72 design around an 84-layer passive midplane, potentially enabling denser AI computing while raising questions about production, cooling and orbital deployment.