Loading video player...
The search landscape is undergoing a significant transformation. Microsoft's retirement of Bing Search API and Google's limitation on results per query have sparked a debate on data accessibility. For years, search APIs like Bing and Google Custom Search have been integral to web development, offering easy access to web results, images, and news. However, the rise of generative AI and retrieval-augmented generation (RAG) has shifted the focus to flexible retrieval layers within the AI pipeline. Microsoft and Google's recent moves indicate a strategic shift towards AI-mediated access within their ecosystems, prioritizing controlled web data for higher-level AI services. This realignment of control has opened doors for new players like Perplexity and Parallel, who are designing search APIs specifically for AI workloads, emphasizing retrieval quality and low latency. The search API market is thriving, with a focus on AI-native infrastructure. Vespa, an open-source engine, plays a crucial role in this new generation of search APIs. As search infrastructure becomes an integral part of the AI stack, the ability to handle structured and unstructured data efficiently becomes a key differentiator. While the large providers take a step back, innovators are pushing forward, rebuilding the web for AI. This video explores the implications of these changes and how they shape the future of web organization and data accessibility.