Polar Raises $5.7M to Automate Knowledge Work

The big picture: Kevin Jiang, who helped build Comet at Perplexity, is now leading Polar, a new startup building an AI browser designed to automate hours of browser-based work. This vision has attracted $5.7 million in seed funding for Polar, aiming to move beyond search and summaries to full workflow execution.

Why it matters:

• Untapped Potential: While AI has enhanced information retrieval and content generation in browsers, a significant portion of knowledge work still involves repetitive, manual browser-based tasks that remain largely unautomated.
• Direct Interaction: Polar’s approach bypasses traditional API integrations, allowing its AI browser to interact with websites by clicking, typing, and navigating just like a human user, enabling automation of complex, multi-step workflows.
• Competitive Edge: The funding arrives as tech giants and startups intensely race to develop advanced AI agents capable of performing real work within existing software, with browser automation emerging as a critical battleground.

How it works:

• Human-like Interaction: Polar’s browser agent mimics human actions, logging into sites, filling forms, and navigating pages using the user’s own accounts to execute tasks across various web applications.
• Long-duration Workflows: Unlike earlier AI browsers focused on summaries or tab organization, Polar is designed to execute entire workflows that can run for minutes or even hours without human intervention, with some tasks reportedly lasting over fifteen hours.
• Benchmark Performance: The company claims its browser agent outperforms systems from OpenAI and Anthropic on browser-agent benchmarks, demonstrating its capability in a highly competitive field.

The catch: The success of Polar’s vision hinges on the reliability and adaptability of its browser agent across the constantly evolving landscape of web applications. As websites frequently update their interfaces and underlying code, maintaining seamless automation without custom integrations presents a significant technical challenge and potential point of failure. Furthermore, the market for AI agents and browser automation is rapidly becoming crowded, with major players like OpenAI, Anthropic, Google, and Microsoft heavily investing in similar capabilities, posing intense competition for adoption and market share.