Why AI agents are the future of web navigation

Don't miss the next shift in web development from platforms to personal agents.

Rachel-Lee Nabors reflects on the web’s shift from an open space to platform-controlled walled gardens. She introduces “enshittification,” Cory Doctorow’s term for how platforms lure users with free services, attract creators with incentives, then prioritize profits at their expense. This fragmentation forces users to navigate multiple restricted platforms. Rachel-Lee argues that relying on browsers from ad-driven companies is outdated and calls for a smarter way to interact with web content.

She envisions AI agents replacing browsers, enabling users to extract, aggregate, and summarize content freely. Technologies like Puppeteer, Playwright, and the Model Context Protocol (MCP) already support this vision. She highlights the revival of RSS-like protocols for decentralized content sharing, offering alternatives to platform-controlled access.

Looking ahead, Rachel-Lee sees a user-driven web where AI-powered agents personalize content. On-device LLMs could enable real-time adaptations, such as text-to-speech or improved readability. She challenges developers to leverage these technologies, arguing that AI removes barriers to information access—the real question is what they will build with it.

Key takeaways

• The web has shifted from open collaboration to platform monopolization, limiting user control over content and data. Rachel-Lee describes the enshittification cycle and how platforms prioritize profit over accessibility and interoperability.
• AI-driven agents offer an alternative to traditional browsers, enabling direct interaction with web content. By leveraging automation, scraping, and open standards like MCP, users can bypass restrictive platforms and aggregate information on their terms.
• The future of the web is personalized, adaptive, and decentralized. Technologies like on-device LLMs and decentralized protocols are creating a more user-friendly, accessible, and customizable internet experience.

Conclusion

Rachel-Lee envisions a future where developers move beyond browser-based interactions and embrace AI-powered agents to navigate and interact with web content. With existing tools already laying the foundation for this shift, she encourages developers to explore new ways of structuring, accessing, and sharing data. The web’s future, she argues, is one where individuals regain control, breaking free from platform constraints to create a more open, adaptable, and intelligent digital experience.

More from We Love Open Source

• Why AI won’t replace developers
• Demystifying external data as a service
• Comparing GitHub Copilot and Codeium
• How to build a multiagent RAG system with Granite
• Build a local AI co-pilot using IBM Granite Code, Ollama, and Continue