HomeArtificial IntelligenceOpenAI acquires Rockset to strengthen its retrieval capabilities

OpenAI acquires Rockset to strengthen its retrieval capabilities

OpenAI is seeking to strengthen its technology stack. The company announced today that it Rock seta provider known for providing a real-time analytics database to support intelligent enterprise applications.

While terms of the deal remain undisclosed, OpenAI has confirmed that it can integrate Rockset's “word class indexing and query capabilities” to strengthen its cross-product query infrastructure. The company also added that Rockset's entire team will move to OpenAI as a part of the acquisition.

This move is OpenAI's second publicly announced major acquisition following its acquisition of New York-based startup Global Illumination, Inc. last 12 months.

The move also comes as competition within the genetic AI space is becoming increasingly intense. Just yesterday, Anthropic, certainly one of OpenAI's biggest competitors, released Claude 3.5 Sonnet, a robust LLM that significantly outperforms OpenAI's recently released GPT-4o in benchmarks.

Even Ilya Sutskever, co-founder and former chief scientist of OpenAI, has founded his own recent AI startup “Safe SuperIntelligence”.

How will Rockset support OpenAI?

Founded in 2016, Rockset provides enterprises with a cloud-based, real-time analytics database that permits developers to construct data-intensive applications at scale, resembling personalization and IT automation.

The product constantly ingests and indexes data from sources resembling Kafka, MongoDB, DynamoDB and S3, enabling real-time information availability and query. It can execute SQL queries on semi-structured data in lower than a second without the necessity for a predefined schema.

At its core, Rockset uses the open source RocksDB persistent key-value store built at Meta (formerly Facebook) as a foundation. It acts as an external secondary index for OLTP databases, data lakes, and streaming platforms, accelerates real-time analytical queries, and provides performance isolation for primary transactional systems. Throughout 2023, the corporate enhanced its product with features aimed toward supporting AI use cases.

With the acquisition by OpenAI, the tech stack created by Rockset, particularly the best-in-class indexing and query capabilities, will now form the premise for OpenAI's retrieval stack.

The corporations haven't yet shared exact details about what this integration will appear like, but one thing is pretty clear: Rockset will enable OpenAI products to reply customer questions with probably the most current and relevant information faster than ever before. Imagine an enterprise-grade GPT that may higher answer user questions with probably the most accurate data.

“Rockset's infrastructure enables corporations to rework their data into actionable intelligence. We are excited to deliver these advantages to our customers by integrating Rockset's foundation into OpenAI products,” said Brad Lightcap, COO at OpenAI, in an announcement.

Companies and startups using AI tools like OpenAI’s models have already tried a way called “Retrieval prolonged generation,“Generative AI,” coined in a 2020 paper by researchers at Meta, University College London and New York University, links generative AI with external knowledge bases, ideally improving the model’s ability to process certain queries and reducing the danger of incorrect answers—resembling answering an worker’s query about expense reporting policies by consulting an organization’s records.

As for Rockset, it stays unclear how the corporate's business with existing customers will change. It has partnered with some notable players within the industry, including Klarna, Meta, Whatnot and Windward. Questions sent by VentureBeat remained unanswered on the time of writing.

“We are excited to partner with OpenAI to empower users, businesses and developers to totally leverage their data by integrating powerful query capabilities into AI,” said Venkat Venkataramani, CEO of Rockset, in an announcement.


Please enter your comment!
Please enter your name here

Must Read