The fundamental architecture of the internet’s search engines, meticulously crafted and refined over decades, was designed with a specific user in mind: humans. These systems were optimized for individuals who require quick, digestible snippets of information, often scanning only the first few results or headlines. However, as artificial intelligence chatbots and autonomous agents increasingly become primary interfaces for navigating the web and executing complex tasks, a critical paradigm shift is underway, demanding a radical re-evaluation of how information is indexed, retrieved, and processed. This evolving landscape has given rise to Keenable, a nascent but ambitious startup that has recently emerged from stealth mode, announcing a substantial $26 million in seed funding to build web search infrastructure tailored specifically for AI.
The Shifting Paradigm: From Human-Centric to AI-Optimized Search
For over two decades, search engines like Google have perfected the art of delivering relevant results to human queries, prioritizing factors like speed, relevance, and user interface for human consumption. This design philosophy, rooted in the limitations of human attention spans and processing capabilities, shaped the very fabric of the web’s discoverability. Yet, the advent of sophisticated AI agents presents an entirely new set of demands. Unlike humans, AI bots can process colossal volumes of information with unprecedented speed and precision, making the current web infrastructure, designed for "ten blue links" and human-readable summaries, inherently inefficient for their needs.
Industry data underscores this shift. Reports from cybersecurity and content delivery network giant Cloudflare have increasingly highlighted a growing share of web traffic attributed to AI crawlers. While traditional search engine bots like Googlebot have always scoured the internet, the rise of specialized AI agents for various applications, from data aggregation to intelligent assistants, indicates a fundamental change in how the web is being accessed and utilized. These AI entities don’t merely scan for keywords; they seek deep contextual understanding, requiring access to larger portions of information, comprehensive grounding data, and robust retrieval mechanisms that go far beyond what a human user might ever need.
Keenable’s Vision: Building the Foundation for AI’s Internet Navigation
At the forefront of addressing this burgeoning challenge are Andrey Styskin, former head of search, AI, and cloud divisions at Russian search giant Yandex, and German AI scientist Matthias Petri. Their collective vision culminated in the founding of Keenable, a company poised to redefine web search for the AI era. The recent $26 million seed funding round, led by Accel and joined by Conviction Partners and several business angels, validates the pressing need for such an infrastructure.
Styskin emphasizes that AI chatbots achieve significantly superior performance when their responses are "grounded" with verifiable source documents. This concept of grounding is pivotal for ensuring factual accuracy, reducing hallucination, and building trust in AI-generated information. "This actually creates a new flywheel that is different from what Google learned from human behavior," Styskin told TechCrunch, highlighting the distinct requirements of AI-driven search compared to its human-oriented predecessor. This "new flywheel" suggests a self-reinforcing cycle where robust, AI-optimized search infrastructure leads to more accurate and capable AI agents, which in turn drives further demand for enhanced data retrieval.
Keenable is not merely conceptualizing this future; it is actively building it. The startup claims to have already constructed a web search index encompassing more than 100 billion documents. Its Application Programming Interface (API) is reportedly in production, actively utilized by several leading AI labs and inference providers during both the critical training phase of AI models and their operational runtime. While the company maintains confidentiality regarding its specific clientele, it recently announced a strategic partnership with voice AI company Gradium, aiming to bolster Gradium’s capabilities with live information retrieval, a testament to Keenable’s practical application in real-world AI scenarios.
Technological Differentiators and Cost Efficiency
Drawing upon his extensive two decades of experience in search engine development at Yandex and Amazon, Styskin elucidates how Keenable’s offering stands apart from conventional enterprise search solutions. Enterprise search, while powerful within an organization’s confined data ecosystem, often proves prohibitively expensive and technically challenging to scale to the vastness of the public internet. "If you do not fine-tune your index structures for a specific task, the cost of serving and scanning the whole internet is enormous because of the volume," Styskin explained. "That’s why you need to innovate on how you can narrow the search space based on your query very fast. This is what we are bringing to the table."
This innovation lies in Keenable’s ability to create highly optimized index structures and retrieval mechanisms that are specifically designed for the unique patterns and demands of AI queries. By efficiently narrowing the search space and leveraging proprietary algorithms, Keenable aims to significantly reduce the computational and financial burden associated with web-scale information retrieval for AI applications. This cost efficiency is a critical factor, especially for AI developers who require vast datasets for training and inference without incurring astronomical operational expenses.
Further enhancing its technological prowess, Keenable is also developing proprietary retrieval capabilities, including an anticipated product named Web Query Language (WQL). This language is designed to empower AI systems to answer complex questions by intelligently synthesizing information from multiple web sources, even when no single document contains the complete answer. This represents a leap beyond simple document retrieval, moving towards sophisticated knowledge synthesis, a capability crucial for advanced AI reasoning and problem-solving.
Strategic Timing and Market Opportunity
Accel partner Zhenya Loginov, who spearheaded the investment in Keenable, highlighted a significant market void that the startup is poised to fill. According to Loginov, AI players currently face a dearth of viable options for web-scale search infrastructure. This scarcity has been exacerbated by strategic shifts from tech giants like Google and Microsoft, who have reportedly taken steps to either shut down or significantly restrict access to their existing search APIs. These moves are widely perceived as efforts to prevent "cannibalization" of their core search businesses and to promote a more "bundled approach" where access to advanced search capabilities is selectively granted to partners within their ecosystems.
For Styskin, these decisions from the incumbents served as a powerful confirmation of the market opportunity he had identified earlier in his career. While working at Amazon alongside Petri on web search infrastructure specifically for AI applications, such as the Alexa voice assistant, he witnessed firsthand the emerging requirements of AI for robust and scalable web access. The Cloudflare data, indicating a substantial and growing proportion of web traffic originating from AI crawlers, further solidified his conviction that a dedicated web search infrastructure built with AI’s needs at its core was not just an advantage, but an impending necessity. This confluence of experience, market observation, and strategic shifts by major players created the perfect storm for Keenable’s inception.
Operational Growth and Navigating the "Innovator’s Dilemma"
With the fresh infusion of capital, Keenable is focused on scaling its operations and expanding its reach. The company currently operates with a lean but highly skilled team of 15 engineering staff distributed across the U.S. and Europe. The immediate plan is to double this headcount by the end of the year, particularly focusing on building out its go-to-market motion and accelerating product development.
However, the path to becoming "the next Google for AI agents" is fraught with challenges, not least of which are the substantial costs associated with building and maintaining a colossal search index. Styskin candidly admits, "Don’t ask – it is painfully expensive." Despite the inherent financial demands, Keenable is committed to meticulous cost management and pacing its growth strategically.
Styskin also invokes the concept of the "innovator’s dilemma," a theory popularized by Clayton Christensen, to explain how even dominant market players can be vulnerable to disruption from smaller, more agile competitors. While acknowledging the "extremely hard" task of persuading users to shift away from Google for general search, Styskin believes that the U.S. giant is potentially "beatable" on "agentic queries" – complex, multi-step queries initiated by AI agents. A smaller, focused company like Keenable, unburdened by legacy infrastructure or the need to protect an existing human-centric search business, can innovate more rapidly and offer a more cost-efficient, purpose-built solution to AI companies. This strategic focus on a niche, albeit rapidly expanding, market segment allows Keenable to challenge incumbents where their current infrastructure may not be optimally aligned with future needs.
The Broader Landscape and Future Implications
Keenable is not alone in recognizing this burgeoning opportunity. Other players have also entered the space, including privacy-focused browser Brave with its search API and AI-native search engine Exa. Furthermore, Google itself is keenly aware of the evolving landscape and is actively overhauling its core search experience to adapt to the AI era, signaling a broader industry-wide recognition of the fundamental shifts at play.
This broader industry movement, encompassing startups and tech giants alike, indicates a shared conviction: whether for humans or for autonomous agents, the long-standing era of the "ten blue links" is undeniably drawing to a close. The future of information access on the web is increasingly becoming a sophisticated interplay of AI understanding, contextual grounding, and highly efficient, purpose-built retrieval systems.
The implications of Keenable’s mission extend far beyond mere technological upgrades. If successful, specialized AI-optimized search infrastructure could:
- Accelerate AI Development: By providing faster, more accurate, and cost-effective access to web-scale data, Keenable could significantly speed up the training and inference processes for advanced AI models, fostering innovation across various AI applications.
- Enhance AI Accuracy and Reliability: Robust grounding with source documents directly addresses critical issues like AI hallucination, leading to more trustworthy and reliable AI agents and chatbots.
- Democratize Advanced AI: By lowering the cost barrier for web-scale data access, Keenable could empower a wider range of AI developers and smaller companies to build sophisticated AI applications, rather than limiting such capabilities to well-resourced tech giants.
- Reshape Human-Computer Interaction: As AI agents become more adept at navigating and synthesizing web information, they will likely transform how humans interact with the internet, moving towards more conversational, task-oriented, and personalized information retrieval experiences.
Keenable’s journey is just beginning, but its strategic positioning, experienced leadership, and significant early funding suggest it is a formidable player in the race to define the future of web search in an increasingly AI-driven world. The challenge is immense, but the potential to reshape how intelligence accesses and understands the vast expanse of human knowledge on the internet is equally profound.
About the Author
