All episodes

    Episode 210 · August 9, 2026 · 6:52

    Cloudflare just gave AI agents keys to the internet

    Cloudflare launched KiteSurf, a new infrastructure layer, to provide AI agents a standardized, secure method for interacting across internet services and platforms. This development addresses a critical "plumbing" challenge, enabling AI agents to move and act at scale without custom connections for every tool. It signals a shift from AI answering questions to AI actively completing tasks on users' behalf.

    Listen to this episode

    Episode breakdown

    What happened

    Cloudflare, an internet infrastructure company handling approximately 20% of all internet traffic, recently introduced a product called KiteSurf. This new offering gives AI agents a standardized and secure way to move between various services and platforms across the internet. Historically, connecting AI agents to different tools and applications required custom-built connections for each interaction, creating a complex and inefficient process.

    KiteSurf aims to simplify this by providing a shared infrastructure layer. Instead of needing a unique engine for every road, KiteSurf allows an AI agent to use a universal connection point, enabling it to interact with multiple platforms seamlessly. This development is crucial for AI agents, which are AI systems designed to perform actions like browsing the web, filling out forms, or sending emails, rather than just answering questions.

    Why it matters

    Cloudflare's launch of KiteSurf represents a significant step in the evolution of AI agents. By standardizing the "plumbing" that allows agents to connect and interact across the internet, KiteSurf removes a major technical hurdle. This move facilitates the scaling and broader deployment of AI agents, enabling them to move beyond answering questions and into actively performing sequences of tasks across different applications and services.

    This infrastructure shift suggests a future where AI handles entire workflows, not just parts of them. For instance, planning a business trip could involve an agent managing flights, calendar updates, team notifications, and client dinner reservations, rather than a user completing each step manually after AI suggests a schedule. Such capabilities could dramatically reduce the "friction" of routine, repetitive tasks that consume significant work time.

    The strategic implication is that organizations and individuals who learn to direct these increasingly capable AI agents effectively will gain a substantial advantage. It shifts the focus from merely understanding AI's technical underpinnings to mastering the clarity of goals and instructions provided to AI systems. This development positions AI agents to "eat that friction for breakfast," transforming how daily operational tasks are executed.

    What to watch next

    • How quickly will other major internet infrastructure providers respond with similar standardized agent connectivity solutions?
    • Which types of businesses and industries will be early adopters of agent-driven workflows enabled by infrastructure like KiteSurf?
    • What new security considerations and standards will emerge as AI agents gain more pervasive access and action capabilities across the internet?
    • How will the development of AI agent instruction and goal-setting tools evolve to match the capabilities provided by platforms like KiteSurf?

    What this means for you

    Business leaders and operators should begin identifying the most repetitive, low-value tasks within their organizations. These include activities like data transfer, email follow-ups, spreadsheet updates, or routine customer inquiries. These tasks represent prime candidates for automation by the next generation of AI agents, which will leverage infrastructure like KiteSurf. Proactively mapping these processes prepares your organization to adopt agent-based solutions as they become more robust.

    Cultivate a culture that prioritizes clear communication of goals and desired outcomes. As AI agents become more prevalent, the ability to effectively direct these systems will be more valuable than technical coding skills. Train teams to think about what they want to accomplish and how an AI agent could be instructed to achieve it, rather than focusing on the mechanics of the agent itself. This clarity ensures that efficient AI systems produce the correct results, not just efficient wrong ones.

    Key takeaways

    • Cloudflare's KiteSurf provides a standardized, secure way for AI agents to operate across internet services.
    • This infrastructure removes a major technical barrier for AI agents, allowing them to connect more easily to various platforms.
    • The development supports a future where AI agents perform entire workflows and tasks on behalf of users.
    • Businesses should identify repetitive tasks for potential automation by capable AI agents.
    • Effective direction and clear goal-setting for AI agents will be a critical skill for future operators.

    FAQ

    What is Cloudflare's KiteSurf and why is it important for AI agents?

    Cloudflare's KiteSurf is a new infrastructure layer designed to give AI agents a standardized, secure method for moving and interacting across different internet services and platforms. This is important because it solves a significant "plumbing" challenge, making it easier for AI agents to connect to various tools without needing custom-built connections for every single interaction. It enables agents to act and connect at scale across the real internet.

    How does KiteSurf change what AI agents can do?

    Before KiteSurf, connecting AI agents to multiple applications was complex, often requiring bespoke engineering for each new interaction. KiteSurf changes this by providing a universal connection point, allowing AI agents to move and act across services more efficiently. This shifts AI's role from primarily answering questions to actively performing multi-step tasks and entire workflows, such as booking a business trip or managing customer inquiries.

    What kind of tasks will AI agents be able to automate with better infrastructure?

    With better infrastructure like KiteSurf, AI agents will be able to automate a wide range of repetitive and friction-filled tasks. This includes moving information between systems, copy-pasting data, sending follow-up emails, updating spreadsheets, filing reports, scheduling calls, and responding to common customer questions. These are tasks that typically consume significant human effort but do not represent core value creation.

    Why is it important for non-engineers to understand these developments in AI agents?

    It is important for non-engineers to understand these developments because AI agents will increasingly handle operational tasks across various industries. The ability to effectively direct these AI systems, by being clear about goals and desired outcomes, will become a crucial skill. Understanding this trend allows individuals and businesses to identify opportunities for automation and prepare for a workplace where AI agents perform more "doing" rather than just "thinking."

    AI AgentsCloudflareAI Infrastructure

    Share with a friend