On May 16, 2012, Google announced the Knowledge Graph in a blog post by Amit Singhal, then a senior vice president of engineering, titled “Introducing the Knowledge Graph: things, not strings.” The phrase captured the shift the system represented: instead of treating a search query as a string of characters to match against web pages, Google would understand it as referring to real-world things, people, places, and concepts, and the relationships among them.
The post explained that the Knowledge Graph let Google disambiguate queries, recognizing, for example, that “taj mahal” could mean the monument, a musician, or a casino, and offer the right interpretation. It could summarize key facts about an entity in a panel beside the search results and surface unexpected but relevant connections, helping users discover information they did not know to ask for. At launch, Google said the graph contained more than 500 million objects and over 3.5 billion facts and relationships between them.
The Knowledge Graph popularized the term “knowledge graph” itself, which has since become standard across the technology industry for structured, entity-centric stores of facts. It also made the entity-and-relationship model visible to ordinary users through the information panels that still appear in search today.
For a business reader, this milestone marks the mainstream arrival of structured knowledge as a product feature. The same idea, representing information as a graph of entities and relationships, now underpins enterprise data integration, virtual assistants, and much of the work on grounding AI systems in verifiable facts.