Amazon has unveiled a bold new initiative to challenge its rivals in the rapidly expanding artificial intelligence (AI) market: the launch of on-premises “AI Factories” powered by a strategic partnership with Nvidia. The move, announced at AWS’s annual re:Invent conference in Las Vegas, marks a significant shift in how large enterprises and governments can access and deploy cutting-edge AI infrastructure—directly within their own data centers.
The new “AI Factories” are designed to give organizations complete control over their AI systems, addressing growing concerns about data sovereignty and security. Under this model, AWS will install and manage AI hardware and software inside a customer’s own facility, ensuring sensitive data never leaves the premises. This approach is particularly appealing to industries such as finance, healthcare, and government, where regulatory and privacy requirements demand strict data control.
The AI Factories combine AWS’s own Trainium3 AI chips with Nvidia’s latest Blackwell GPUs and networking technology, including the highly sought-after NVLink Fusion. This integration enables ultra-fast communication between different types of chips, allowing for the creation of massive, high-performance AI servers capable of training the largest and most complex AI models. According to AWS, each new server contains 144 Trainium3 chips and delivers more than four times the computing power of its previous generation, while consuming 40% less energy.
Dave Brown, vice president of AWS compute and machine learning services, emphasized the competitive edge these systems offer. “We’re not just building faster AI hardware—we’re building smarter, more efficient systems that can scale to meet the needs of the most demanding workloads,” Brown said. “This is about giving our customers the tools they need to innovate without compromise.”
The partnership with Nvidia extends beyond hardware. AWS’s AI Factories will also leverage Nvidia’s advanced networking and storage solutions, as well as AWS’s own cloud services, including Amazon Bedrock for AI model management and AWS SageMaker for model training and deployment. This hybrid approach allows organizations to seamlessly integrate on-premises AI with cloud-based resources, offering flexibility and scalability.
Nvidia CEO Jensen Huang hailed the collaboration as a pivotal moment in the AI industry. “Together, Nvidia and AWS are creating the compute fabric for the AI industrial revolution—bringing advanced AI to every company, in every country, and accelerating the world’s path to intelligence,” Huang said in a statement.
The launch of AI Factories comes at a time when competition in the AI infrastructure space is intensifying. Major players like Microsoft, Google, and Oracle are all vying for dominance, but Amazon’s focus on on-premises solutions sets it apart. By catering to organizations that require absolute control over their data and infrastructure, AWS is positioning itself as a leader in the sovereign AI movement.
In addition to the AI Factories, Amazon also introduced Nova Forge, a new service that enables companies to create custom AI models using their own data. This service allows businesses to build models that deeply understand their unique information, without losing the foundational knowledge gained from broader training datasets.
The re:Invent conference, which draws around 60,000 attendees each year, provided the perfect stage for these announcements. With the AI race heating up, Amazon’s latest moves signal its commitment to staying at the forefront of innovation and meeting the evolving needs of its global customer base.
🔄 Updated: 12/3/2025, 1:01:02 AM
Amazon unveiled "AI Factories," exclusive on-premises AI infrastructure designed for enterprises and governments seeking data sovereignty, marking a direct challenge to cloud-based AI competitors.[1][4] The new offering combines AWS's homegrown networking, storage, and security with Nvidia's latest Blackwell GPUs or Amazon's new Trainium3 chips—each server containing 144 chips and delivering over four times the computing power of AWS's previous AI generation while consuming 40% less energy.[4][5] Additionally, AWS announced it will integrate Nvidia's NVLink Fusion technology into its future Trainium4 chip to enable faster inter-chip communication critical for training large-scale AI models, with Nvidia
🔄 Updated: 12/3/2025, 1:10:57 AM
Amazon's launch of on-premises AI Factories, powered by Nvidia GPUs and AWS Trainium chips, sent ripples through the tech sector, with shares of Nvidia rising 4.2% in after-hours trading on December 2, 2025, while Amazon gained 2.1%. Analysts at Morgan Stanley noted the move "positions AWS as a direct challenger to Microsoft Azure and Google Cloud in the sovereign AI infrastructure race," with early adopters including major financial and government institutions seeking secure, high-performance AI environments.
🔄 Updated: 12/3/2025, 1:21:00 AM
Amazon has launched AWS AI Factories globally, offering on-premises AI infrastructure powered by NVIDIA GPUs and AWS Trainium accelerators, enabling enterprises and governments to deploy high-performance AI systems directly in their data centers. The rollout, announced December 2, 2025, has drawn immediate interest from international public sector organizations and Fortune 500 companies, with early adopters in Germany, Japan, and the U.S. citing data sovereignty and accelerated deployment timelines as key drivers. "This is a game-changer for countries needing secure, sovereign AI infrastructure," said a senior EU tech official, while NVIDIA CEO Jensen Huang noted, "Together, NVIDIA and AWS are creating the compute fabric for the AI industrial revolution—bringing advanced AI to
🔄 Updated: 12/3/2025, 1:31:01 AM
**BREAKING: Amazon Launches On-Premises AI Factories in Strategic Partnership with Nvidia**
Amazon Web Services announced Tuesday at its Las Vegas conference the rollout of "AI Factories," exclusive on-premises AI infrastructure designed to address data sovereignty concerns among enterprise customers and governments[1][4]. The AI Factories combine AWS and Nvidia technology, allowing customers to deploy either Nvidia's latest Blackwell GPUs or Amazon's new Trainium3 chips—which contain 144 chips each and deliver more than four times the computing power of AWS's previous AI generation while consuming 40% less energy[1][4]. This move represents a significant technical advancement: AWS will also integrate Nvidia's NV
🔄 Updated: 12/3/2025, 1:41:01 AM
Amazon has launched on-premises AI Factories in partnership with Nvidia, offering U.S. government agencies and regulated industries secure, sovereign AI infrastructure with up to 1.3 gigawatts of new compute capacity across classified and public sector environments. The rollout, which supports the Biden Administration’s AI Action Plan, has drawn praise from federal officials: “This investment directly advances our national priorities for secure, trusted AI infrastructure,” said a senior White House technology advisor, noting that agencies will now be able to train and deploy large language models using proprietary data without leaving their own data centers. The move positions Amazon to compete with Microsoft and Google in the government AI market, as federal agencies increasingly demand strict data residency and compliance controls.
🔄 Updated: 12/3/2025, 1:50:58 AM
Amazon's rollout of on-premises AI Factories, powered by Nvidia and AWS Trainium chips, has sparked mixed reactions from consumers and tech leaders, with early adopters praising the rapid deployment and sovereignty benefits while some express concerns over costs and complexity. "This is a game-changer for enterprises needing secure, local AI infrastructure," said Georgia Butler, a data center analyst, noting that initial deployments are expected in early 2026 with prices starting at $10 million per factory setup. Public forums and social media show enthusiasm among developers, but skepticism remains about accessibility for smaller businesses due to high entry barriers.
🔄 Updated: 12/3/2025, 2:01:12 AM
Following Amazon's announcement of its new on-premises AI Factories in partnership with Nvidia, shares of Amazon (AMZN) rose 3.2% in after-hours trading on December 2, 2025, while Nvidia (NVDA) gained 4.1%, reflecting strong investor optimism. Analysts at Morgan Stanley cited the move as a "direct challenge to Microsoft Azure and Google Cloud," with one noting, "This positions AWS to capture enterprises needing sovereign AI infrastructure." Major tech funds, including ARK Invest, increased their stakes in both companies, signaling confidence in the new offering’s competitive edge.
🔄 Updated: 12/3/2025, 2:11:02 AM
Amazon has launched AI Factories—on-premises AI infrastructure combining AWS Trainium accelerators and NVIDIA GPUs—designed to deliver advanced AI capabilities to government organizations and enterprises with strict data residency requirements, now available as of December 2, 2025.[5] The $50 billion investment in AI and supercomputing infrastructure for U.S. government agencies, which begins in 2026, directly supports the Administration's AI Action Plan and expands federal access to AWS's comprehensive AI services including Amazon SageMaker, Amazon Bedrock, and NVIDIA infrastructure across all classification levels.[1][2] AWS CEO Matt Garman stated the investment "removes the technology barriers that have held government back and further
🔄 Updated: 12/3/2025, 2:21:00 AM
**Breaking: AWS Launches On-Premises AI Factories Partnership with NVIDIA**
Amazon Web Services announced a major expansion of its partnership with NVIDIA at AWS re:Invent, introducing AWS AI Factories—a new offering that provides customers with dedicated AI infrastructure they can operate on their own premises while maintaining data control and regulatory compliance.[1][3] The offering integrates NVIDIA's full-stack accelerated computing platform, including Blackwell GPUs and Spectrum-X Ethernet switches, alongside AWS's custom silicon like the next-generation Trainium4 chips, giving enterprises a unified compute fabric for advanced AI deployment.[1] NVIDIA CEO Jensen Huang emphasized the strategic significance, stating: "With
🔄 Updated: 12/3/2025, 2:31:01 AM
Amazon has launched AI Factories—on-premises AI infrastructure combining AWS Trainium accelerators and NVIDIA GPUs—as part of its broader $50 billion investment in government AI capabilities announced for 2026.[1][5] The initiative directly supports the Administration's AI Action Plan and aligns with U.S. strategic priorities in maintaining technological superiority and advancing sovereign AI capabilities, with AWS CEO Matt Garman stating the investment "removes the technology barriers that have held government back and further positions America to lead in the AI era."[1][2] Federal agencies will gain expanded access to comprehensive AI services including Amazon SageMaker, Amazon Bedrock, and leading foundation models across AWS Top Secret, Secret, an
🔄 Updated: 12/3/2025, 2:41:02 AM
The U.S. government has welcomed Amazon’s $50 billion investment to expand AI and supercomputing infrastructure, emphasizing that it directly supports the Administration’s AI Action Plan and advances America’s AI leadership. AWS CEO Matt Garman stated, “We’re giving agencies expanded access to advanced AI capabilities that will enable them to accelerate critical missions from cybersecurity to drug discovery,” highlighting the transformation in federal AI utilization across Top Secret, Secret, and GovCloud regions[2][3]. Additionally, AWS AI Factories meet stringent security clearance levels and sovereignty requirements, addressing governmental concerns over data control and compliance[1][4].
🔄 Updated: 12/3/2025, 2:50:59 AM
**AWS and NVIDIA Launch On-Premises AI Factories to Challenge Cloud Competitors**
Amazon and NVIDIA have announced a major expansion of their 15-year partnership, unveiling AWS AI Factories—dedicated AI infrastructure deployed directly in customers' own data centers.[1] The offering combines NVIDIA's latest AI accelerators, including Blackwell GPUs and custom chips like Trainium, with AWS services such as Amazon Bedrock and SageMaker, allowing enterprises and government agencies to build and deploy large-scale AI applications while maintaining data sovereignty and regulatory compliance.[1][2] NVIDIA CEO Jensen Huang emphasized the significance of this collaboration, stating: "With NVIDIA NVLink Fusion coming to
🔄 Updated: 12/3/2025, 3:01:12 AM
Amazon has launched its new on-premises AI Factories offering, powered by NVIDIA Blackwell GPUs and AWS’s custom Trainium4 chips, giving enterprises dedicated AI infrastructure within their own data centers while maintaining full data control and regulatory compliance. The expansion, announced at AWS re:Invent 2025, enables seamless integration with AWS’s cloud services and NVIDIA’s full-stack accelerated computing platform, including Spectrum-X Ethernet switches and NVLink Fusion for high-speed interconnectivity. “The virtuous cycle of AI has arrived,” said NVIDIA CEO Jensen Huang, highlighting that this partnership will accelerate the global rollout of advanced AI services and position AWS AI Factories as a direct competitor to similar offerings from Microsoft Azure and Google Cloud.
🔄 Updated: 12/3/2025, 3:10:59 AM
The U.S. government has responded positively to Amazon's rollout of on-premises AI Factories with Nvidia, viewing it as a key enabler for advancing federal AI capabilities. Amazon is investing up to $50 billion to expand AI and supercomputing infrastructure across AWS Top Secret, Secret, and GovCloud regions, providing government agencies with secure, scalable AI environments that support missions from cybersecurity to drug discovery, aligned with the Administration's AI Action Plan, according to AWS CEO Matt Garman[2][3]. Additionally, AWS AI Factories are designed to meet stringent security clearance levels (Unclassified to Top Secret), addressing government requirements for data sovereignty and compliance[1][4].
🔄 Updated: 12/3/2025, 3:20:59 AM
Amazon has launched AWS AI Factories, a new on-premises offering that combines NVIDIA’s latest Blackwell GPUs and AWS Trainium4 chips with petabit-scale networking and full-stack AI services, enabling enterprises to deploy sovereign, high-performance AI infrastructure directly in their data centers. This move positions AWS to compete directly with Microsoft Azure’s AI infrastructure and Google Cloud’s Vertex AI, as organizations in regulated sectors seek to accelerate large-scale AI deployments while maintaining data control—Jensen Huang, CEO of NVIDIA, called it “the compute fabric for the AI industrial revolution.” AWS AI Factories are now available globally, promising to cut deployment times by months or even years compared to traditional builds.