Global semiconductor revenues are estimated to double from 2024 to 2030 to reach over $1 trillion, driven by AI transformation to build infrastructure and consumption end points for GenAI to Agentic AI to Physical AI applications.
This is according to new research from Counterpoint Research’s AI 360 practice titled What’s Next for AI?: How Will It Shape the Semiconductor Market? The research group says the primary catalyst will be the development of advanced AI server infrastructure, driven by sustained and potentially rapid growth in demand for forthcoming AI applications. Most of this is currently coming from hyperscalers in both the short and longer term, supporting GenAI’s path to maturity across text and, increasingly, audio and visual in the near term and Physical AI in the longer term along.
Neil Shah, Counterpoint Vice President of Research, has this to say: We are witnessing the rise of ‘token economy,’ as the token becomes the new AI currency. The first phase of infrastructure rollout drove an increase in token consumption as the market moved from basic text-based applications to richer, multi-modal GenAI combining text, image, audio, and video.
The ongoing second phase is supporting an exponential increase in token generation for Agentic AI applications – from complex conversational AI and semantic search to fully integrated multimedia content creation. This wave will demand significant compute, memory, and networking capabilities, both in the cloud and at the edge, and the impact on semiconductor consumption cannot be overstated.
The third wave will support the advent of Physical AI, enabling the rise of autonomous machines such as humanoids, industrial robots, and vehicles. This is tremendous news for the semiconductor industry, which has been laying the foundation to enable advanced AI at scale and transform lives, industries, and economies.
