Urgent.News

the world's headlines, one feed

AI

Cloudflare wants to provide the operating system for the AI-first enterprise

Traditional operating systems (OS) were built to manage hardware, files, apps, and users on a device, but Cloudflare says the agentic AI era requires a whole new format. The company this week announced Cloudflare OS , which connects AI agents, enterprise data and context, internal systems, and workflows together in one secure workspace. It is open source and browser-based, sparing companies the…

Cloudflare wants to provide the operating system for the AI-first enterprise

Cloudflare has unveiled Cloudflare OS, an operating system tailored for the AI-first enterprise landscape. This open-source, browser-based workspace aims to connect AI agents, enterprise data, context, internal systems, and workflows into a unified, secure environment. Rita Kozlov, VP of product at Cloudflare, emphasized that Cloudflare OS is not a conventional desktop OS but rather a reimagination of workplace computing for AI.

The platform is accessible directly through a browser and is integrated into an enterprise's Cloudflare account, enabling users to perform tasks such as researching, creating documents, building applications, and automating workflows without requiring a terminal.

Cloudflare OS is designed to be highly customizable, allowing organizations to configure their own access policies, models, branding, skills, and integrations. The system starts with zero permissions for AI agents, which are only granted access to the specific tools necessary for a given task. This granular control is facilitated through govern-managed connectors called "gatekeepers," which enable administrators to manage what data AI agents can access, modify, or require human approval for.

Additionally, Cloudflare OS tracks the resources needed by each agent, ensuring that appropriate access controls follow the agent's work when shared.

The company initially developed Cloudflare OS for internal use, with employees across various departments utilizing it daily. Over the past 30 days, internal users have reportedly created more than 4,000 apps, automations, and tools. Cloudflare's sales team has also reportedly saved an estimated 10,000 hours by automating previously manual tasks, such as territory planning and proposal creation.

Cloudflare has positioned Cloudflare OS as a cohesive bundle of tools, offering a single platform for enterprises to manage their AI-related workflows and infrastructure. This approach contrasts with traditional operating systems like Windows, macOS, or Linux, and even competitors like Microsoft and Google, which have not fully integrated their AI offerings under a unified brand.

According to tech analyst Carmi Levy, Cloudflare OS is "more cohesively bundled" and infrastructure-focused, making it easier for enterprise IT buyers to understand and incorporate AI-related platforms and workflows into their existing systems.

As AI introduces new architectural requirements, Cloudflare OS addresses these challenges by serving as a single-source vendor for AI-related infrastructure. This infrastructure-first, application-agnostic approach allows Cloudflare OS to coexist with existing AI applications from different providers, such as OpenAI, Anthropic, Google, Microsoft, or Meta.

The open-source nature of Cloudflare OS also minimizes the risk of vendor lock-in, as organizations can inspect, customize, and connect their own systems, ensuring that the platform remains truly theirs.

Written by urgent.news from Computerworld's reporting — not their text. Machine-written; read the original for the full account.

Read the original at computerworld.com →

More in AI

I Built an AI Minecraft Thumbnail Maker for YouTube Creators

Creating a Minecraft video is already a lot of work, and thumbnails can easily become another slow step. A good thumbnail needs a clear story, a strong focal point, readable text, and the right…

  • AI tool simplifies Minecraft-themed YouTube thumbnail creation.
  • Minecraft-specific formats and customization options available.
  • High-resolution outputs and feedback encouraged for improvement.

The Silent Costs of AI APIs Nobody Warns You About

I remember the exact moment the excitement turned to dread. I had just integrated GPT-4 into a side project—a small document summarization tool.

  • Token imbalance doubles costs, with GPT-4 charging double for output tokens.
  • System prompt tax adds unseen input token charges, even unused context.
  • Retry costs and rate limits create hidden fees, especially with scaling.