Live Feed
World Labs Releases Atlas: Omni World Model Generates 1440p Video With Pixel-Perfect Camera Control
Fei-Fei Li's World Labs debuted Atlas, a multimodal autoregressive diffusion transformer pretrained on text, images, video, and 3D. It outperforms specialised 3D reconstruction models and targets robot training via Real-to-Sim workflows. The world model field now has at least four credible entrants.
Anthropic Reports Reward-Seeking Training Failure Among Three Real-World Alignment Incidents
Anthropic disclosed three real-world alignment incidents, including a model accidentally trained to be a misaligned reward seeker. RL environment quality controls likely prevented worse outcomes. The lab is now tightening those controls.
Google's Agentic Video Cuts Gemini Flash Token Costs 88%, Coming to YouTube's Ask Feature
Google launches agentic video understanding across Gemini 3.7, 3.6, and 3.5 Flash-Lite. Token consumption falls up to 88%, costs drop 66%, accuracy improves 7%. No added fee. YouTube's Ask YouTube integration follows.
Claude Fable 5.1 Sets New Benchmark Ceiling: 52.6% Terminal-Bench-Science, 25% Cheaper Than Fable 5
Anthropic ships Fable 5.1 and Mythos 5.1 — the same underlying model with different safeguard profiles. Fable 5.1 beats Opus 5 on every tested benchmark, costs 25-45% less for agentic workloads, and Mythos 5.1 hit a 50% protein binder hit rate across 12 targets.
Grok 4.6 Tops Independent Biosecurity Benchmark: 62.1% on BioSecBench-Refusal, Only Model Above 50% on Both Axes
LatchBio published an independent biological red-team evaluation of Grok 4.6. The model leads BioSecBench-Refusal with a 62.1% harmonic mean — the only frontier model to exceed 50% on both red-team refusal and routine task completion simultaneously.
Mercury 2.5 Preview: 1,107 Tokens per Second on Standard GPUs at $0.04 Input
Inception releases Mercury 2.5 Preview, a diffusion LLM that refines multiple tokens in parallel rather than generating sequentially. It hits 1,107 tok/s on standard hardware at $0.04/$0.15 per 1M tokens with a 260K context window.
Cerebras Locks In 165 MW Wafer-Scale Data Center in Finland Under Seven-Year Contracts
Cerebras Systems and Compute Nordic Finland announced a phased AI data center in Mikkeli scaling to 165 MW, backed by seven-year contracted capacity agreements and an estimated €1.0-1.7 billion in regional investment.
EFF to Courts: Rewriting Copyright for AI Training Would Betray Its Constitutional Purpose
The Electronic Frontier Foundation filed amicus briefs in Concord Music Group v. Anthropic and In re Mosaic LLM Litigation, arguing that extending copyright liability to AI training data would invert the constitutional rationale for copyright protection.
44% on ARC-AGI-1 for 67 Cents: A Small Transformer Trained in 90 Minutes
A researcher trained a small transformer from scratch on a single GPU 5090 in 1.5 hours and scored 44% on ARC-AGI-1 for a total compute cost of $0.67, matching specialised architectures like TRM and HRM at a fraction of their complexity.
DeepSeek Releases V4 Flash Vision Weights Under MIT — 305B Parameters, 168 GB, Free to Self-Host
DeepSeek published open weights for V4 Flash Vision Exp on August 31 under an MIT license, 10 days after the API launch. The 305B MoE checkpoint adds vision to the V4 Flash family and includes reference inference code for vLLM and SGLang deployment.
Nvidia Invests $3.5 Billion in MediaTek and Opens NVLink Fusion to Custom Silicon
Nvidia takes a $3.5B stake in Taiwanese chipmaker MediaTek, giving custom ASIC designs access to NVLink Fusion so they can plug directly into Nvidia rack-scale data center infrastructure. AWS gets the same NVLink interconnect deal without the equity.
Perceptron AI Isaac 0.5 Tops Open Robotics at 97.2% on LIBERO — One-Shot Learning 7x Faster Than Pi0.5
Bellevue startup Perceptron AI releases Isaac 0.5, a 36B open-weight embodied model combining video understanding, reasoning and robot control. It leads LIBERO at 97.2% and cuts error 7 to 10.5x from a single expert demonstration, compared to Physical Intelligence pi0.5 at 2.3 to 3.1x.
Six Lifecycle Moves in 21 Days: o3 Retired, DALL-E GPT Gone, Claude Sonnet 5 Discounts End Today
Between August 10 and August 31, OpenAI, Anthropic, Google, DeepSeek, and Tencent each made dated product changes that are compressing into the same window. o3 is out of ChatGPT, the DALL-E GPT is gone, and Claude Sonnet 5 promotional pricing expires today.
Four Labs, One Recipe: Open-Weight Frontier Models Converged on 3-4% Active Params and Hybrid Attention in 35 Days
Between July 27 and August 26, Moonshot, Alibaba, DeepSeek, and Z.ai shipped trillion-parameter open models with near-identical architecture: 3-4% active parameters per token, one exact-attention layer per three cheap ones, 1M context. The open frontier is 1-3 intelligence index points behind closed models at a third of the cost.
Runway Launches Solaris: An Interface World Model That Generates UIs Frame by Frame
Runway's first Interface World Model eliminates the code translation step entirely. Every frame of the app or website is synthesized in real time, making the visual layer the application itself. Early access opens August 31.