AI Apple quietly registers genai.apple.com subdomain ahead of WWDC26 Top Stories
Pinned News
AI Apple quietly registers genai.apple.com subdomain ahead of WWDC26 
Developer Context Is a Budget — Not a Trash Can 
OpenAI Codex: More Than a Coding Assistant, It's a Work Automation System 
Anthropic Why Claude Code Is Ditching Markdown for HTML 
Anthropic Elon Musk shuts down xAI, hands its 220,000 GPUs to Claude 
AI Musk's lawyers may have just blown open the xAI-OpenAI probe 
OpenAI The decade of OpenAI: ambition, conflict, and betrayal News,Images,Videos
Most Recent
xAI Has 500,000 GPUs but Uses Only 11% of Them. Rivals Say That's 'Ridiculously
xAI's massive GPU cluster runs at just 11% utilization — a figure one rival calls 'ridiculously low.' The problem isn't unique to Musk's company.
Cursor's new SDK lets you spawn coding agents anywhere — including your customers' apps
Cursor just dropped a TypeScript SDK that lets you spin up the same programming agent that powers its desktop app — in your CI/CD pipeline, on a cloud VM, or even inside your own product. A few lines of code is all it...
Zhipu AI Traces GLM-5 Gibberish to Race Conditions in KV Cache
Zhipu AI explains the root cause of garbled output in its GLM-5 coding agent model: two race conditions in the KV cache, plus a trick for anomaly detection using speculative decoding acceptance rates.
Grok's new Imagine Agent turns one sentence into a whole short film on an
xAI is testing a new agent mode for Grok that swaps the chat interface for an infinite canvas, letting you generate complete short films, comic strips, or product media from a single prompt.
Zed 1.0 arrives: the code editor that ditched Electron for Rust and a GPU
After five years of development, the Atom founder's new editor Zed hits 1.0 with a GPU-accelerated UI, multi-agent AI features, and a plan to let humans and AIs edit code together in real time.
Alibaba's QoderWake AI agent cuts root cause analysis from 30 minutes to 2
Alibaba's new AI agent QoderWake slashes root cause analysis from 30 minutes to 2 minutes, using a memory-driven architecture that learns from every task.
White House blocks Anthropic's Mythos expansion over compute crunch
The White House rejected Anthropic's plan to let 70 more companies use its vulnerability-finding AI Mythos, citing security and compute shortages — partly because the government wants to run it itself.
OpenAI's goblin problem traced to a single reward signal gone rogue
OpenAI traced the GPT models' obsession with goblin metaphors to a reward signal meant for the 'Nerdy' personality. The signal contaminated training data, causing goblin mentions to spike 175%.
NVIDIA's Blackwell is 35x cheaper per token — if you enable the right settings
NVIDIA's new Blackwell GPUs cost nearly double per hour to rent, but a 65x boost in token throughput slashes per-million-token cost from $4.20 to $0.12 — at least with all software optimizations turned on.
Nvidia's China ban fuels Cambricon's explosive Q1 growth
Cambricon's Q1 revenue doubled and profit nearly tripled, driven by AI demand and US export controls that have cut off Nvidia and AMD from the Chinese market.