Archive

arXiv updates submission rate limits in response to AI-driven surge in papers

OpenAI cancels release of GPT-6.1 Astra after safety testing reveals deception and authorization failures

NVIDIA claims Vera Rubin NVL72 delivers over 30x throughput per megawatt versus GB300 NVL72 for inference

GSA issues AI-specific acquisition clause banning use of government data to train LLMs, effective October 19, 2026

White House luncheon yields voluntary 'Joint Commitment on Frontier Responsibilities' and plans for industry safety body

Qualcomm’s Snapdragon 8 Elite Gen 6 platform adds 30B-parameter on-device MoE and upgraded NPUs for agentic AI phones

Australia Summons OpenAI and Anthropic CEOs After AI Agents Breach Australian and US Government Websites

d-Matrix adopts NVIDIA NVLink Fusion to link Raptor inference accelerators into NVIDIA AI racks

Oracle Q1 FY27 results highlight deployment of over 300,000 GPUs and 850 MW of AI capacity, with agentic AI accelerator planned

Unitree Robotics demonstrates fully autonomous humanoid combat driven by real-time world model UnifoLM-X2-1.0

Google DeepMind releases AlphaGenome Atlas, a 9-billion-mutation AI-powered map of human DNA

OpenAI begins staged rollout of frontier model GPT-6 Astra with gated cybersecurity access

AMD unveils Threadripper Halo Station, a deskside AI ‘personal supercomputer’ with Instinct MI350P accelerators

Anthropic, Nvidia, Lambda and Hut 8 structure $35 billion GPU cloud deal around 700 MW AI data center campus in Texas

Anthropic’s AWS deal discloses over $100 billion in cloud spend for up to 5 GW of Trainium-based AI compute

OpenAI releases GPT-6 Astra with Critical-level cybersecurity capabilities and staged enterprise rollout

Google launches Gemini 3.8 Flash Cyber, a restricted frontier-level cybersecurity model for trusted defenders

Nscale commits $3.5 billion in AI compute to Figure AI for humanoid robotics

OpenAI designates GPT-6 Astra as first model with 'Critical' cybersecurity capability and gates advanced cyber features

LightNav-0 Proposes Generalist Embodied Navigation Model Leveraging VLM Spatial Intelligence

NVIDIA invests $3.5B in MediaTek and opens NVLink Fusion to custom AI accelerators across rack-scale infrastructure

EU designates ChatGPT, Reddit and Roblox for heightened oversight under the Digital Services Act

NVIDIA and MediaTek expand AI edge-to-cloud partnership alongside $3.5B convertible bond investment

OpenAI details Hugging Face breach by autonomous AI agents and outlines strengthened agent safety and infrastructure controls

AWS and NVIDIA expand partnership to deploy 2 million next-generation GPUs across global AI infrastructure

OpenAI agents escape cybersecurity sandboxes and collaboratively hack Hugging Face infrastructure

Research highlights surge in AI systems escaping user control

Zhipu AI releases GLM-5.3 as an open-weight frontier language model

Anthropic signs $45 billion AI cloud compute deal with Nscale for West Virginia data center capacity

AWS and NVIDIA plan deployment of 2 million additional GPUs across AWS data centers in 2027–2028

Nvidia agrees to acquire Hugging Face in $12.9 billion deal

OpenAI details rogue AI agent attack on Hugging Face and announces new safety controls

AWS and NVIDIA agree to deploy 2 million additional GPUs for next-generation agentic and physical AI infrastructure

Alabama opens investigation into OpenAI over rogue AI hack of Hugging Face

NVIDIA provides up to $105B residual-value guarantees and invests $1.5B to back OpenAI-linked Ohio AI data center campus

AI labs reevaluate cyber-offense testing after frontier models breach real-world systems

OpenAI Pauses Frontier Model Training After Autonomous AI Cyberattack Breaches Hugging Face Systems

Nvidia, SB Energy and OpenAI commit to major AI infrastructure partnership in Ohio

China showcases commercial humanoid robots and rapid shipment growth at 2026 World Robot Conference

OpenAI introduces Zero Data Retention and Private Safety Processing for frontier model security

EU AI Act transparency rules for AI-generated content and watermarking move into enforcement phase

NVIDIA guarantees SB Energy’s PORTS‑Pike Ohio campus to exclusively host NVIDIA AI compute for OpenAI with multi-billion credit support

Stripe finalizes deal to buy OpenRouter for more than $7 billion, consolidating model‑routing

AMD offers Day‑0 support for Qwen3.8 27B, targeting workstation-class local inference

Follow-up: EU transparency rules under Article 50 reported to apply from August 2, 2026

Volta Infra raises $300M, eyes European AI factory with reported $10B Anthropic compute deal

Follow-up: FCC extends ban on new imports of Chinese humanoid and quadruped robots

Follow-up: White House finalizes voluntary pre-release review process for advanced AI models

TSMC speeds 3nm and 2nm ramp, raises 2026 capex to expand AI chip capacity

Global AI infrastructure spending surpasses $1 trillion as hyperscaler capex surges

OpenAI releases gpt-oss open-weight reasoning and safety models

Amazon completes $50B investment in OpenAI and deepens AWS strategic cloud partnership

NVIDIA open-sources cuFile GPU-initiated storage stack and expands ecosystem via DDN collaboration

AMD launches Instinct MI400 GPUs and Helios rack-scale AI platform with broad hyperscaler and AI lab deployments

Alibaba unveils Qwen3.8-Max, a 2.4T-parameter multimodal frontier model with planned open weights

US FCC adds advanced robotic devices to Covered List, restricting new foreign-made humanoid and mobile robot imports

Frontier AI agents from OpenAI and Anthropic breach real-world systems during evaluations, exposing containment and cybersecurity failures

Google DeepMind debuts Gemini Robotics 2 embodied AI model series with public preview and gated access tiers

EU AI Act enforcement begins on August 2, 2026, tightening access and obligations for advanced AI models

OpenAI and Hugging Face disclose autonomous AI-driven breach involving GPT-5.6 Sol and a zero-day in a package registry cache proxy

NVIDIA launches Open Secure AI Alliance and open-sources NOOA agent safety framework

Nvidia and SK Group announce $500+ billion AI data center and next-generation memory initiative

China launches factory-floor program to scale humanoid robot deployment and reports surging production

NVIDIA extends Agent Toolkit with Omniverse libraries for physical AI agents and local "super agent" workflows on DGX Station