Rubin

• Nvidia CEO Jensen Huang announced expectations of at least $1 trillion in demand for its Blackwell and Rubin AI systems through 2027, doubling from $500 billion projected through 2026. • This projection signals over 100% annual sales growth for Nvidia amid booming data center demand. • The forecast underscores AI's role in driving U.S. tech sector expansion and GDP growth, with analysts citing non-inflationary productivity gains.

• NVIDIA CEO Jensen Huang announced at GTC in San Jose a forecast of $1 trillion in sales from Blackwell and Rubin chips by late 2027, doubling the prior $500 billion estimate through 2026. • The company unveiled a new inference system generating 700 million tokens per second, 350 times faster than the Hopper generation, to counter custom chips from competitors like Google. • NVIDIA's DRIVE platform for robotaxis, valued at $1.2 trillion by Morgan Stanley, is adopted by Uber and BYD, with Uber's fleet launching in 2028; data center revenue hit $192 billion last year, up 66%.

• Nvidia announced at GTC 2026 that it has locked in $1 trillion in orders for its Blackwell and Vera Rubin chips through 2027, far exceeding annual revenue and signaling massive AI infrastructure investment across the industry. • Major hyperscalers including Amazon, Microsoft, and Google have significantly ramped up investments in AI data centers and infrastructure, competing to build massive language models and prepare for exponential growth in AI workloads. • Blackwell chips have already begun shipping while Vera Rubin is positioned to deliver major boosts to AI training and inference performance, addressing the current AI infrastructure race among tech giants.

NVIDIA announced the Rubin platform, named after astronomer Vera Rubin, marking a shift from Blackwell with extreme co-design across six new chips targeting 1.6nm processes for massive computing power gains. First Rubin systems will launch in the second half of 2026 through AWS, Microsoft, and Google, with Microsoft integrating them into 'Fairwater' AI superfactories. This hardware leap supports edge-centric AI via specialized ASICs, enabling real-time insights, amid rising demand projected over $700 billion in global datacenter leases.

Microsoft is incorporating rack-scale NVIDIA Rubin systems into its 'Fairwater' AI superfactories to power advanced model training starting late 2026. This follows NVIDIA's Rubin unveiling, promising leaps in compute efficiency for cloud-scale AI. The move bolsters US AI infrastructure amid enterprise agent adoption rates surging in datacenters.