Nitro Studio Ai
The latest Nitro Studio Ai coverage — news, analysis, and updates from the WindowsNews.AI desk.
Beyond the Hype: Google Gemini 2.5 Pro vs. OpenAI’s GPT-5 – Choosing the Right AI for Your Workflow
Google Gemini 2.5 Pro and OpenAI's GPT-5 represent two distinct AI philosophies: Gemini tightly integrates into Google Workspace with large-context reasoning and built-in video generation, while ChatGPT excels in conversational polish and platform neutrality with clearer data opt-outs. Our analysis breaks down pricing, privacy, and performance to help you pick the right assistant for research, coding, and creative work.
Elon Musk’s xAI Files ‘Macrohard’ Trademark, Outlines AI Software Factory Fueled by Grok Ads
Elon Musk’s xAI has filed a trademark for ‘Macrohard,’ signaling plans to build an AI-driven software factory powered by agentic workflows and advertising in Grok. While the move pressures Microsoft and excites some developers, enterprise trust gaps, IP concerns, and energy constraints could impede adoption. Windows professionals should monitor developments but enforce strict governance around AI-generated code.
xAI Drops Grok Code Fast 1: Agentic Coding Now in VS Code at $1.50/M Output Tokens
xAI has launched Grok Code Fast 1, an agentic coding model optimized for low-latency tool calls and multi-step engineering tasks inside IDEs like Visual Studio Code. With introductory pricing of $0.20 per million input tokens and $1.50 per million output tokens, the model targets continuous agentic development, but its benchmark claims remain unverified. Enterprise adoption requires careful governance to mitigate security, licensing, and compliance risks.
How the AGI Clause, GPT-5, and Open Weights Turned the Microsoft-OpenAI Partnership Into a High-Stakes Rivalry
OpenAI and Microsoft are navigating a complex new phase, balancing co-dependence with rivalry. The AGI clause stands as a pivotal contractual lever, while both companies pursue independent AI strategies. Enterprises must prepare for multi-cloud flexibility and shifting vendor dynamics.
MIT's VaxSeer AI Outperforms WHO in Forecasting Dominant Flu Strains, Retrospective Study Shows
MIT's VaxSeer AI combines protein language models, epidemiological simulations, and antigenicity prediction to forecast dominant influenza strains. In retrospective tests, it outperformed WHO recommendations for the notoriously difficult H3N2 subtype in nine out of ten seasons and matched or exceeded for H1N1. While promising, integration into global vaccine decision-making requires prospective validation, transparency, and new governance protocols.
AI Chatbots Now Train on Your Conversations by Default – Here’s How to Protect Your Privacy
Anthropic’s switch to default opt-in model training on consumer chats has spotlighted how Claude, ChatGPT, and Gemini now require users to actively opt out to prevent their conversations from being used for AI improvement. Despite available toggles, long retention windows, opaque human review, and dark-pattern consent designs leave real privacy risks—practical steps and regulatory pressure are needed to close the gap.
Windows 11 Beta Build 26120.5770 Rolls Out One-Click Excel Table Conversion and Braille Viewer
Windows 11 Build 26120.5770 adds a Click to Do action that converts on-screen tables into Excel spreadsheets with one click, plus a Narrator Braille viewer and AI profile cards. Features require a Copilot+ PC and Microsoft 365 subscription, and come with known issues like Bluetooth Xbox controller crashes. The Beta build also includes taskbar, audio, and HDR fixes.
Microsoft’s AI Sustainability Tools Gain Traction as EU Simplifies ESG Rules, but Data Governance Risks Remain
European regulators are streamlining sustainability reporting standards while cracking down on greenwashing, pushing companies toward auditable metrics. Microsoft’s Cloud for Sustainability and AI integrations offer automation, but data lineage and dual‑use cloud governance risks demand immediate attention from IT and legal teams.
Vivaldi Rejects AI as Microsoft, Google, and Anthropic Push Cloud and Data Collection
A cascade of policy changes and product updates this week underscored the growing tension between user privacy and the convenience of cloud-connected AI. Anthropic reversed its privacy stance to train Claude on chats by default, Microsoft made Word's default save location the cloud, Google expanded data collection in Play and Translate, and Samsung TVs gained Copilot—while Vivaldi’s CEO publicly rejected embedding AI in its browser. Users and admins must now actively adjust settings to protect data, as defaults shift toward greater platform control.
Syracuse U. Libraries Tackle AI Ethics, Copilot Security, and Big Data in Fall Workshops
Syracuse University Libraries’ Fall 2025 free workshop series teaches AI literacy, secure Microsoft Copilot use, and ProQuest TDM Studio for big data research, while emphasizing data privacy, citation integrity, and Windows-specific best practices.
Windows 11 25H2 Hits Release Preview as Tiny Enablement Package—Most Updated PCs Already Have It
Windows 11 25H2 has arrived in the Release Preview channel as a minimal enablement package, meaning the update is already pre-staged on most up-to-date PCs. It introduces Start menu improvements, AI-powered search and actions, Quick Machine Recovery, and removals of legacy components, while IT teams must validate compatibility and new feature gating.
Windows 11 Dev Channel 26220.5770 Packs Excel Table Conversion and Braille Viewer—With Strings Attached
Windows 11 Dev Channel build 26220.5770 introduces Click to Do’s Excel table conversion, inline Microsoft 365 profile cards, and a Narrator Braille viewer, but the features are gated behind hardware, regional, and licensing restrictions. The update also brings bug fixes and a smarter share UI, though known issues remain.
xAI Drops Grok Code Fast 1: A Speedy, Tool-Savvy AI Coder Priced to Disrupt
xAI has launched Grok Code Fast 1, an agentic coding model designed for low-latency, tool-driven development with aggressive per-token pricing. Available through partners like GitHub Copilot with a free trial, it scores 70.8% on SWE-Bench-Verified while security and privacy concerns demand cautious adoption.