The evolution of AI: From AlphaGo to AI agents, physical AI, and beyond
In March 2016, the world witnessed a unique moment in the evolution of artificial intelligence (AI) when AlphaGo, an AI developed by DeepMind, played against Lee Sedol, one of the greatest Go players of the modern era. The match reached a critical juncture in Game 2 with Move 37, where AlphaGo made a move so unconventional and creative that it stunned both the audience and Lee Sedol himself.
This moment has since been recognized as a pivotal point in the evolution of AI. It was not merely a demonstration of AI’s proficiency in playing Go but a revelation that machines could think outside the box and exhibit creativity. This moment fundamentally altered the perception of AI, transforming it from a tool that follows predefined rules to an entity capable of innovation. Since that fateful match, AI continues to drive profound changes across industries, from content recommendations to fraud detection. However, the game-changing power of AI became evident when ChatGPT brought generative AI to the masses.
Experience the latest in AI innovation. Join Microsoft at the NVIDIA GTC AI Conference. Learn more and register.
The critical moment of ChatGPT
The release of ChatGPT by OpenAI in November 2022 marked another significant milestone in the evolution of AI. ChatGPT, a large language model capable of generating human-like text, demonstrated the potential of AI to understand and generate natural language. This capability opened up new possibilities for AI applications, from customer service to content creation. The world responded to ChatGPT with a mix of awe and excitement, recognizing the potential of AI to transform how humans communicate and interact with technology to enhance our lives.
The rise of agentic AI
Today, the rise of agentic AI — systems capable of advanced reasoning and task execution — is revolutionizing the way organizations operate. Agentic AI systems are designed to pursue complex goals with autonomy and predictability. They are productivity enablers that can effectively incorporate humans in the loop via the use of multi-modality. These systems can take goal-directed actions with minimal human oversight, make contextual decisions, and dynamically adjust plans based on changing conditions.
Deploy agentic AI today
Microsoft and NVIDIA are at the forefront of developing and deploying agentic AI systems, providing the necessary infrastructure and tools to enable advanced capabilities such as:
Azure AI services: Microsoft Azure AI services have been instrumental in creating agentic AI systems. For instance, the Azure AI Foundry and Azure OpenAI Service provide the foundational tools for building AI agents that can autonomously perceive, decide, and act in pursuit of specific goals. These services enable the development of AI systems that go beyond simple task execution to more complex, multi-step processes.
AI agents and agentic AI systems: Microsoft has developed various AI agents that automate and execute business processes, working alongside or on behalf of individuals and organizations. These agents, accessible via Microsoft Copilot Studio, Azure AI, or GitHub, are designed to autonomously perceive, decide, and act, adapting to new circumstances and conditions. For example, the mobile data recorder (MDR) copilot at BMW, powered by Azure AI, allows engineers to chat with the interface using natural language, converting conversations into technical insights.
Multi-agent systems: Microsoft’s research and development in multi-agent AI systems have led to the creation of modular, collaborative agents that can dynamically adapt to different tasks. These systems are designed to work together seamlessly, enhancing overall performance and efficiency. For example, Magnetic-One, a high-performance generalist agentic system, is designed to solve open-ended tasks across various domains, representing a significant advancement in agent technology.
Collaboration with NVIDIA: Microsoft and NVIDIA have collaborated deeply across the entire technology stack, including Azure accelerated instances equipped with NVIDIA GPUs. This enables users to develop agentic AI applications by leveraging NVIDIA GPUs alongside NVIDIA NIM models and NeMo microservices across their selected Azure services, such as Azure Machine Learning, Azure Kubernetes Service, or Azure Virtual Machines. Furthermore, NVIDIA NeMo microservices offer capabilities to support the creation and ongoing enhancement of agentic AI applications.
Physical AI and beyond
Looking ahead, the next wave in AI development is physical AI, powered by AI models that can understand and engage with our world and generate their actions based on advanced sensory input. Physical AI will enable a new frontier of digitalization for heavy industries, delivering more intelligence and autonomy to the world’s warehouses and factories, and driving major advancements in autonomous transportation. The NVIDIA Omniverse development platform is available on Microsoft Azure to enable developers to build advanced physical AI, simulation, and digital twin applications that accelerate industrial digitalization.
As AI continues to evolve, it promises to bring even more profound changes to our world. The journey that was sparked from a single move on a Go board to the emergence of agentic and physical AI underscores the incredible potential of AI to innovate, transform industries, and elevate our daily lives.
Experience the latest in AI innovation at NVIDIA GTC
Discover cutting-edge AI solutions from Microsoft and NVIDIA, that push the boundaries of innovation. Join Microsoft at the NVIDIA GTC AI Conference from March 17 to 21, 2025, in San Jose, California, or virtually.
Visit Microsoft at booth #514 to connect with Azure and NVIDIA AI experts and explore the latest AI technology and hardware. Attend Microsoft’s sessions to learn about Azure’s comprehensive AI platform and accelerate your innovation journey.
Learn more and register today.
This content was produced by Microsoft and NVIDIA. It was not written by MIT Technology Review’s editorial staff.