AI startup Keenable has secured $26 million in seed funding led by Accel to develop a web search index specifically for AI agents. The company aims to move beyond human-centric search, focusing on providing accurate data for AI systems. Investors in the tech space are tracking this, as it addresses the growing demand for independent search infrastructure amid rising restrictions from major search providers.
Keenable, a startup founded by former executives from Yandex and Amazon, has raised $26 million in a seed funding round led by Accel, with participation from Conviction Partners and other investors. The company is developing a new web search index optimized not for human users, but for artificial intelligence agents. This technology aims to solve a significant problem in the current AI landscape: the need for AI systems to accurately verify their information against real-time, source-based web data.
Moving Toward AI-First Infrastructure
Traditional search engines are built for humans, relying on keywords and ranking systems designed to capture human attention. However, AI agents require a different approach. They need direct access to raw information to 'ground' their responses, ensuring they do not hallucinate or provide incorrect data. Keenable has already built a web index containing over 100 billion documents to facilitate this process. The company is also developing a proprietary tool called WebQueryLanguage, which helps AI systems synthesize information from multiple web sources more effectively.
For investors following the AI infrastructure sector, this move highlights the growing importance of independent data access. Major search providers like Google and Microsoft have recently started restricting access to their search APIs. This creates a market gap for independent infrastructure providers that can offer reliable, large-scale search capabilities at a manageable cost. By building its own index, Keenable aims to bypass these restrictions and provide a service that is both cost-efficient and tailored for machine interaction.
Technical and Financial Risks
While the concept is promising, building a web-scale index is an expensive and technically demanding task. The cost of crawling, indexing, and maintaining up-to-date information for the entire internet is extremely high. Keenable’s success will depend on its ability to manage these infrastructure costs while scaling its operations. The company plans to use the new funding to double its engineering team by the end of the year to accelerate this development.
Competition is another factor to monitor. Keenable must compete with existing players like Brave and Exa, as well as the internal search initiatives of major technology companies. Furthermore, the reliance on demand from AI labs and inference providers means the company’s growth is tied to the broader adoption of AI agents. If the demand for these specific AI services does not scale as expected, or if incumbent tech giants find ways to lower their own costs or improve their APIs, the business model could face pressure. The next important step for the company will be proving the commercial viability of its API and successfully scaling its operations to support more enterprise-grade AI applications.