Nvidia CEO Jensen Huang at GTC 2024 presented the company’s highly anticipated Blackwell graphics processing unit (GPU).
Mr Huang said: “Generative AI is the defining technology of our time. Blackwell GPUs are the engine to power this new industrial revolution. Working with the most dynamic companies in the world, we will realize the promise of AI for every industry.”
Undoubtedly, there are challenges and competitors in that up-and-coming market niche.
Competitors AMD (AMD) and Intel (INTC) aren’t sitting idly by; they are indeed working on their chips, hoping to catch Nvidia eventually.
Nvidia customers, including Microsoft (MSFT), Amazon (AMZN), Google (GOOG, GOOGL), Meta (META), and Tesla (TSLA), are currently using or actively developing their in-house AI chips as alternatives to Nvidia’s offerings. Part of the reason is that they don’t have to pay the tens of thousands of dollars Nvidia’s chips are estimated to cost.
Mr. Huang's response to that expected challenge is strong, intelligent and innovative.
In addition to the Blackwell and Grace Blackwell chips, Nvidia also debuted a chain of three technical components DGX – OVX - AGX:
DGX SuperPOD is a supercomputer system.
OVX is a concept of Omniverse Digital Twins for design validation, testing and forecasting.
AGX is a platform and special chips optimised for sensors and robots.
That is the strong part of the response.
The second part of the response is the intelligent and innovative concept of an AI factory.
The AI Factory comprises three main pillars:
Technology for manufacturing AI models, running, training, and packaging them for customers.
Tools to modify AI models.
Infrastructure to run AI models. The models can be deployed on the Nvidia cloud, other clouds or on-premise.
The models will be supplied in a container called Nvidia Inference Microservice (NIM), a Pre-trained AI model, and packaged and optimised to run on a NEMO platform.
The AI Factory is a new part of Nvidia's business strategy. With that step, Nvidia will take the strategic position on the downstream frontline and work directly with the end customers and, potentially, with consumers.
The services offered include manufacturing pre-trained AI models, hosting and support services, and hardware production, forming a solid ecosystem and comprehensive value chain in the AI industry.
Nvidia's advantage in the manufacturing AI platform allows it to take the lead and leverage the ecosystem synergy effect to advance more rapidly, leaving competitors uncertain where to focus their efforts.
The AI Factory will significantly impact and change other industries and business sectors not directly involved in the AI niche competition.
The factory offers quick and cost-efficient solutions for integrating AI technologies into other businesses. That opportunity is not only about intelligent chatbots.
The leading optimisation will be in operation processes, enterprise entire operation models, operational quality and operation cost.
Companies which integrate AI into their processes correctly and avoid pitfalls and mistakes will gain a competitive advantage. Companies which waste time will be pushed out of the market.
That positive and exciting picture is missing one crucial component. AI transformation will be more complex, challenging, and mandatory than, for example, digital transformation, which is the former headline and hot theme.
There must be a service which will help enterprises, businesses, and organisations prepare and integrate AI models into operations.
AI transformation will be more complex, challenging, and MANDATORY.
It will be mandatory for all companies which compete without budget or monopoly advantages. It will be compulsory for states which plan to stay abreast with geopolitical competitors.
IT transformation is not about funny or cool androids working around and helping with shopping, cleaning or nursing.
AI transformation is about better management, data processing, and new management practices and methods.
Comments