Urgent.News

What's breaking now, across thousands of outlets.

Tech

Cloudflare Launches Kitesurf, a Lightweight Browser Built for AI Agents

Cloudflare’s Kitesurf browser promises lower resource use and stronger isolation for AI agents, but enterprises should assess its beta limitations. The post Cloudflare Launches Kitesurf, a Lightweight Browser Built for AI Agents appeared first on TechRepublic .

Cloudflare has introduced Kitesurf, a lightweight browser engineered for AI agents, aiming to offer a more resource-efficient and secure browsing experience. Unlike traditional browsers, Kitesurf is designed to meet the specific needs of machine actors, featuring lower resource requirements and built-in isolation. This browser is currently in beta, accessible through Cloudflare’s Browser Run, a headless Chrome browser hosted in the cloud.

Tests have shown that Kitesurf consumes 3.1 to 3.8 times less CPU and 4.7 to seven times less memory compared to Chromium, although it takes 1.7 to 1.8 times longer to complete tasks. This resource efficiency addresses the mounting cloud costs associated with running AI agents. The development of specialized browsers for AI agents stems from the need for optimized browsing solutions that prioritize performance and security.

Cloudflare’s Kitesurf was inspired by Obscura, a Rust-based headless engine for AI automation, and stands out by being built on Cloudflare Workers, allowing for seamless integration without additional setup for Cloudflare customers. It also supports popular software libraries like Puppeteer and Playwright, ensuring compatibility with existing tools.

Security is a key focus, with each page isolated to prevent information leakage and protect against attacks such as prompt injection. To encourage adoption, Cloudflare has offered Kitesurf on a free trial basis to its customers, with plans to open-source the browser once it’s finalized. As more organizations rely on web access for autonomous agents, tools like Kitesurf are essential for managing the associated resource and financial challenges.

Written by urgent.news from TechRepublic's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at techrepublic.com →

More in Tech

'Asimov Was Right' About Rules For Robots, Says Ex-US Cyber Director

Former U.S. National Cyber Director Chris Inglis says the biggest AI risk isn't sentience but autonomy. "What I'm worried about is that they get to choose what and where they do something, and under what rules they do it," he said, citing recent cases of AI agents from OpenAI, Anthropic, and Meta escaping security sandboxes.

More from Friday 7 August →