Anthropic confidentially filed its S-1 with the SEC and raised $65 billion in Series H at a $965 billion post-money valuation, led by Altimeter Capital, Dragoneer, and Sequoia. The company is preparing for a potential IPO.

Cybersecurity company Cyera is closing a $300 million funding round led by Evolution Equity Partners at a $12 billion valuation. The round reflects explosive growth in the AI-related security sector.

Uber capped employee AI spending after budget depletion in just 4 months. The adjustment follows active encouragement for staff to use AI tools without restrictions.

NVIDIA and Microsoft announced a collaboration to deliver a complete platform for deploying agentic AI from Windows devices to Azure cloud. The initiative includes optimized hardware, secure runtimes, and models tuned for extended reasoning tasks.

Anthropic expands Project Glasswing to 150 organizations across 15+ countries, focusing on critical infrastructure in power, water, healthcare, and communications. The program has already identified over 10,000 critical security vulnerabilities in code audits.

OpenAI expanded Codex capabilities with new plugins and tools aimed at white-collar work for analysts, marketers, designers, and investors. The company released an internal report on Codex usage in knowledge work.

Travelers launched an AI-powered Claims Assistant using OpenAI that guides customers through filing claims and provides 24/7 support. The solution enables the insurer to scale operations during peak demand periods.

NVIDIA introduced NemoClaw, a tool automating industrial engineering workflows beyond simulations, including design, meshing, debugging and reporting. More than a dozen industrial software leaders joined the initiative.

Microsoft launched Scout at Build 2026, a personal assistant integrating OpenClaw-style capabilities into Microsoft 365. The assistant automates productivity tasks directly within Outlook, Teams, and OneDrive.

Anthropic expands Project Glasswing with 150 new organizations across 15+ countries to audit and secure critical infrastructure. The program has detected over 10,000 severe vulnerabilities since its initial launch.

Microsoft announced new AI models, agentic tools similar to OpenClaw, and Windows updates at its annual developer conference. The event showcased Microsoft's focus on AI for developers.

Google demonstrated Gemini Spark, an AI agent for trip planning that impressed and raised concerns about its capabilities. The experience shows advances in complex multi-step agents.

OpenAI released new Codex plugins, sites, and annotations helping multidisciplinary teams become more productive with AI. The tools integrate into existing workflows.

Opal, famous for its premium webcam, pivoted to consumer electronics with backing from OpenAI and Samsung investments. Its first product will be an AI-powered audio gadget.

Google added a feature to its Phone app that detects fraudulent calls impersonating known contacts using AI. The system automatically flags suspicious calls to protect users from identity impersonation scams.

Microsoft unveiled Surface RTX Spark Dev Box, a mini developer PC based on Nvidia's Arm-based RTX Spark chips. The device is optimized for sustained workloads and local AI tasks.

Microsoft launched MAI-Thinking-1 at Build 2026, marking an ambitious step into in-house model development. The announcement reflects Microsoft's strategy to reduce OpenAI dependency after renegotiating their partnership agreement.

Hugging Face introduced Holo3.1, a model for fast local AI agents that runs without remote server dependency. Ideal for use cases requiring privacy and low latency.

The AI Alliance (led by IBM and Meta) unveiled Project Tapestry, exploring how to build frontier-scale models through a decentralized global coalition. The initiative aims to balance cutting-edge capability with local sovereignty and data control.

Financial institutions are adopting Transaction Foundation Models to unify fragmented AI systems (fraud, credit, risk). This approach enables better understanding of customer financial behavior.

President Trump signed an executive order creating a voluntary framework for companies to share frontier models with the federal government before release. The measure aims to promote secure innovation and strengthen critical infrastructure cybersecurity.

After industry pressure, Trump signed a revised executive order requiring only voluntary pre-release government reviews of advanced AI models. The approach is less restrictive than the initial version.

NVIDIA announced JetPack 7.2 and NemoClaw support on Jetson, enabling agentic AI capabilities on edge devices. The update includes CUDA 13, performance improvements, and Multi-Instance GPU support.

Microsoft released ASSERT, an open source framework allowing developers to create AI evaluations using natural language descriptions. It significantly simplifies behavior testing for AI models.

Microsoft just kicked off Build 2026 with a keynote from CEO Satya Nadella and other company leaders. As expected, it was filled with announcements, ranging from new Surface hardware to an always-on p

Much like Google, Microsoft is launching its own version of OpenClaw. Microsoft Scout is an always-on assistant that integrates into Microsoft 365 apps like Outlook, OneDrive, and Microsoft Teams, all

Microsoft’s OpenClaw-style agent appears in Teams, just like a human colleague, and automates your dull office tasks.

submitted by/u/ThereWas[link][comments]

Hey there! I wrote a blogpost about my experience running local models on a V100 from a newbie perspective and got loads of views outside of reddit, so I thought I'd share it here too!submitted by/u/t

ZeroDrift launched a compliance service that sits between AI models and end users to flag and replace problematic messages. The solution protects against regulatory compliance risks.