Why Block Secured NVIDIA's Latest AI System in North America
Block, Inc., the payments and financial services company behind Square and Cash App, announced it will become the first North American enterprise to implement NVIDIA's DGX SuperPOD with DGX GB200 systems.
The supercomputing installation at an Equinix data centre will support Block's research into open source generative AI models focused on applications for financial inclusion.
The move represents a computing infrastructure investment for Block as it expands its artificial intelligence capabilities beyond previous work in deepfake detection and generated audio technology.
As the first North American company to secure this specific NVIDIA technology, Block establishes itself at the forefront of financial services firms investing in specialist AI hardware.
The deployment comes as competition intensifies among financial technology companies to secure advanced computing hardware amid growing demand for AI development resources in the financial sector.
Technical infrastructure designed for frontier model development
The NVIDIA Grace Blackwell DGX SuperPOD systems that Block has secured provide specialised computing architecture for training and running large AI models.
These systems feature NVIDIA's newest graphics processing unit (GPU) technology, designed to address computational limitations that have constrained development of more sophisticated AI systems.
“With NVIDIA DGX GB200 systems, Block engineering and research teams can develop frontier open-source AI models that can tackle complex, real-world challenges with state-of-the-art AI supercomputing”
The GB200 systems incorporate both CPU and GPU capabilities on the same hardware, which reduces latency in AI processing.
“The industry, and the world, is undergoing a seismic shift with adopting AI tools.
"At Block, we think it's essential not only to apply AI to existing problems, but also to explore, learn and build in the open so that we can advance the frontier of AI in a way that truly levels the playing field for our customers and community,” says Dhanji R. Prasanna, CTO of Block.
Block tested its requirements through Lambda 1-Click Clusters, a service that provides temporary access to interconnected NVIDIA GPUs supported by machine learning engineers.
This validation step allowed Block to confirm its technical hypotheses before committing to a full deployment. Lambda now offers these clusters with NVIDIA's new Blackwell architecture.
"At Block, we think it's essential not only to apply AI to existing problems, but also to explore, learn and build in the open so that we can advance the frontier of AI"
Charlie Boyle, Vice President of DGX platforms at NVIDIA, notes the growing demands of sophisticated AI development: “As AI models grow in complexity and scale, businesses need powerful infrastructure that can match the pace of innovation.
“With NVIDIA DGX GB200 systems, Block engineering and research teams can develop frontier open-source AI models that can tackle complex, real-world challenges with state-of-the-art AI supercomputing.”
Strategic data centre partnership for AI deployment
Block has selected Equinix, a data centre and digital infrastructure company, to host the new AI systems.
Equinix specialises in providing facilities with advanced cooling capabilities, high-performance networking and compliance with data privacy regulations.
The data centre environment offers low-latency connections to cloud providers and other technology partners, an important consideration for AI workloads that require substantial data throughput and processing power.
Jon Lin, Chief Business Officer at Equinix, explains the infrastructure requirements: "Frontier models represent the cutting edge of artificial intelligence technology, pushing the boundaries of what AI can achieve, and they require the latest in AI chips - like NVIDIA's new DGX SuperPOD.
“By deploying at Equinix's neutral, cloud-adjacent platform, companies like Block can unlock expanded compute scale and flexibility. This enables the customisation of AI solutions with a choice of infrastructure, cloud, models and cooling at our neutral exchange.”
Open source approach to AI development
Block's investment in advanced computing aligns with its recent introduction of an open source AI framework codenamed “goose”.
This framework connects large language models (LLMs) – AI systems trained on vast text datasets that can generate human-like text responses – to applications that perform real-world actions.
The company reports that while initial applications of goose focus on software engineering tasks, both internal teams and external developers are exploring additional use cases beyond engineering.
Block's approach stands in contrast to many financial institutions that develop proprietary AI systems.
“Block is a technology company first, and engineering is at the heart of everything we do"
The open source strategy could potentially accelerate innovation in financial technology by allowing external developers to build on Block's research.
The company's portfolio includes Square for merchant payment processing, Cash App for consumer money transfers, Afterpay for instalment payments, TIDAL for music streaming and cryptocurrency ventures including mining services.
This diverse range of financial services provides Block with multiple potential applications for its AI research.
Dhanji emphasises the company's technical foundations and previous experience with AI systems: “Block is a technology company first, and engineering is at the heart of everything we do.
“Our past work on generative speech models means Block is well-positioned for innovation surrounding Gen AI models, and all possible applications that come with it.”
Explore the latest edition of FinTech Magazine and be part of the conversation at our global conference series, FinTech LIVE.
Discover all our upcoming events and secure your tickets today.
FinTech Magazine is a BizClik brand
