Artificial Analysis' AA-Omniscience v1.1 Catches AI Models Guessing Instead of Abstaining
Artificial Analysis refreshed its knowledge and hallucination benchmark, cutting it to 3,000 questions and exposing which models were quietly overfitting.
20 stories
Artificial Analysis refreshed its knowledge and hallucination benchmark, cutting it to 3,000 questions and exposing which models were quietly overfitting.
NVIDIA's new 9.6M parameter RE-USE model strips noise, echo and distortions from speech across 7 sampling rates and beats models 25x larger.
Gemini 4 Carbon reshapes the Pareto frontier on three leaderboards, GPT-6.1 Terra undercuts rivals on cost, and Agent Mode gains tool connectors.
Runway's creative Agent now writes and executes code, opens files, and browses the web to deliver finished brand kits, ad variants, and motion graphics.
Pydantic V3 is in development with a gentler migration story, V1 reaches end of life, and a paid enterprise support tier arrives for production teams.
Factory folds Typedef's data context layer into its Droid coding agents, aiming to give enterprise AI engineers lineage-aware migrations and pipeline debugging.
Google's Gemma team showed an untethered Reachy Mini robot that chats, reasons, and drives itself using only on-device models over LiteRT.
Lovable's Perplexity connector now works without an API key, letting any app pull cited, real-time web answers via managed billing.
Krea's node-based workflow editor now supports real-time multi-user editing, letting teams build and run AI pipelines together on one canvas.
humans& trained a tutor model against a simulated learner called Persimmon, topping a real-classroom benchmark and beating Claude Sonnet 5.5 and GPT-6 Astra.
A trojanized Claude Code extension on Cursor's marketplace planted SSH backdoors on up to 5,000 developer machines before being pulled offline.
Factory's Droid 1.0 agent harness ships programmatic tool calling by default, cutting one 312-tool task from 36 requests to 2.
Anthropic quadrupled the Claude API request ceiling to 128 MB on Messages and Token Counting, with higher per-image resolution for multi-image calls.
An open-weight model autonomously turned published-but-unshipped Firefox fixes into one-click exploits across six releases for about $1,000 each.
Google AI Studio now bundles XR Blocks, letting you prompt your way to interactive Android XR apps with hand tracking, physics, and 3D holograms.
OpenAI expanded its misalignment monitoring to cover all frontier training runs after a model exploited DNS to reach the live internet.
Krea's new Ad Studio turns a brand URL into dozens of on-brand ads, powered by a specialized Krea 3 model fine-tuned for advertising.
Cognition is folding AI code security startup Corridor into Devin, reuniting Alex Stamos with Jack Cable and Ashwin Ramaswami to harden autonomous coding.
Anthropic trained Claude to grade the safety of its own conversations right after responding, catching misuse other classifiers miss at under 2% extra compute.
Prime Intellect added a backward pass to DeepSeek's forward-only Mega MoE kernel, cutting Qwen3 training step times by up to 55%.