Browser Use, a groundbreaking company founded by Magnus Müller and Gregor Zunic, has successfully raised $17 million in a seed funding round. The company graduated out of ETH Zurich’s Student Project House accelerator. Yet it has rapidly drawn outsized interest from the investor and developer community. Astarcia Myers from Felicis was instrumental in convincing her partners to lead us into seed funding. Famed investors including Paul Graham, A Capital, and Nexus Venture Partners participated.
Browser Use is all about making the world wide web more readable for AI bots. This improvement provides them access to the internet and makes it easier for them to interact with web pages. Using web scraping in tandem with a data-driven approach, the platform empowers browsers to cut through the noise and get to work quickly. Clearly, many companies are bullish on this unconventional approach. With a little luck, over 20 other businesses in this most recent Y Combinator Winter batch have fully adopted Browser Use to their model already! The firm is a member of Y Combinator’s 2025 winter batch.
“We convert [websites] into something agents can understand. This approach means we can run the same tasks again and again at a cheaper cost,” said Müller.
The impact of Browser Use stretches well beyond that specific aspect. For developers, it’s being positioned to become an important layer that makes it possible for AI agents to more easily and gracefully interact with websites. Use cases such as that envisioned by Chinese startup Butterfly Effect have already demonstrated this potential. They leveraged Browser Use in their widely shared Manus viral tool. The platform’s open-source-first philosophy makes it even more attractive and accessible.
Astar’s Myers, who led the round, is excited about web AI agents.
“We think web AI agents are the next frontier that really helps with the end-to-end automation of human tasks,” Myers stated.
Chrome Browser Use solves some of the biggest problems AI agents face when navigating a website. Especially for dynamic websites that frequently change their structure, this can be really tricky for AI agents. Müller recognized this problem, pointing to examples such as LinkedIn where agents usually do not succeed because the environment is always shifting.