Sakhanda Wire
NVDA $155.46 -0.44% MSFT $506.24 -2.27% GOOGL $280.20 -1.24% META $618.87 -1.20% AMZN $257.89 +0.05%

Latest news

Older stories

Ars Technica

FAA expands AI air traffic advice across US airspace

The Federal Aviation Administration is expanding an artificial-intelligence system that advises air traffic controllers to all 22 air traffic control centers across the United States. The agency says the move is intended to help reduce delays and fuel use while improving the experience of airline passengers. The system, called SMART, was initially deployed in airspace over Washington, DC, on September 21.

TechCrunch

Muse opens business calls to paid U.S. subscribers

Meta’s consumer AI agent Muse can now place calls to businesses in the United States for users, following an initial test in September. The feature is available only to paid subscribers. Meta said Muse had surpassed 10 million users last month, as the company works to position consumer AI as a tool for routine tasks.

The Verge

Google introduces Gemini 4 Hydrogen for more complex questions

Google says its new Gemini 4 Hydrogen model is intended to provide more helpful answers when users submit complex questions. The company specifically includes queries with several parts, difficult-to-understand typos, and highly specific requests among the cases the model is designed to handle. According to Google, Gemini 4 Hydrogen can break a question into separate elements and respond to each of them.

MarkTechPost

CopilotKit releases self-hostable OpenBot 1.0 with policy controls

CopilotKit has released OpenBot 1.0, an MIT-licensed workspace for running AI agents under a company’s own control. It can be used locally or deployed through a Helm chart on AWS EKS, Azure AKS, Google Cloud GKE, or another Kubernetes cluster. Teams provide their own agents, models and infrastructure rather than using a fixed vendor environment.

TechCrunch

Nebius plans $40 billion AI infrastructure spending by 2027

Nebius plans to spend $40 billion on AI infrastructure by 2027, a commitment far above Wall Street expectations. The neocloud company intends to finance much of the expansion through customer prepayments and loans secured by customer contracts. Prepayments and asset-backed financing for investment-grade agreements could cover up to 85% of an expected $40 billion to $45 billion in 2027 capital expenditure.

The Decoder

Anthropic brings Claude into video calls with Claude Meetings

Anthropic has launched Claude Meetings, a feature that adds Claude to video calls as a visible participant. During a session, the AI can answer questions, create documents and whiteboards, and record notes and action items. It responds in text by default and can use connected tools when users grant permission.

TechCrunch

Reuters reports a third alleged affair involving Jensen Huang

Reuters reported that Jensen Huang, Nvidia’s founder, had a previously undisclosed relationship with another company employee. Reuters has not identified the employee, but reported that she works for Nvidia. The report adds to earlier accounts involving Huang and Alice Huang, who has been accused of spying, as well as Nvidia Vice President of Health Care Kimberly Powell.

MIT Tech Review

Nvidia told lawmaker chip tracking to deter smuggling is technically feasible

Representatives of Nvidia privately told Illinois Representative Bill Foster that the company could technically implement measures to track advanced AI chips, Foster told MIT Technology Review. The measures are included in the proposed Chip Security Act, which would use geolocation checks to help prevent chips from being smuggled to China. Nvidia has not confirmed this publicly, and its representatives did not provide Foster with a written statement.

The Decoder

Gemini 4 Hydrogen outperforms the larger Argon in benchmarks

Google has released Gemini 4 Hydrogen, a general-purpose AI model for everyday coding, research and agentic workflows. It is positioned as a successor to Gemini 3.8 Flash, while Google’s frontier models Gemini 4 Carbon and Gemini 4 Argon target more demanding, long-running reasoning tasks. Hydrogen offers low, medium, high and xhigh thinking levels, plus a one-million-token context window for both input and output. API pricing is $1.50 per million input tokens and $7.50 per million output tokens; cached input costs $0.075.

The Verge

Google researcher’s AI-led Maya site claim challenged

Earlier this week, Google researcher Pietro Miotti said a team of AI agents had identified a potentially unmapped Maya site in Mexico. In a blog post, Miotti described reviewing the agents’ reports and seeing villages and other regular shapes that appeared to resemble structures or houses.

TechCrunch

Google upgrades Search with Gemini 4 Hydrogen

Google is rolling out Gemini 4 Hydrogen in Search, aiming to improve how AI Mode handles complicated questions. The model is designed to interpret multi-part requests, highly specific prompts and confusing typos, while responding more quickly. Google describes Hydrogen as its biggest model improvement to Search so far.

Ars Technica

Google rolls out Gemini 4 Hydrogen to replace 3.8 Flash

Google has begun rolling out Gemini 4 Hydrogen, the newest model in its elemental naming series. The system is intended to replace the faster and more economical Flash models used across Google’s ecosystem, and it has taken the place of Gemini 3.8 Flash in the company’s products.