SoftBank’s Strategic Investments in AI and Semiconductor Sectors SoftBank has recently intensified its investments in artificial intelligence (AI) and semiconductor assets, focusing on enhancing U.S. and global technological infrastructure. Notably, the company has committed $2 billion in equity to Intel, signaling strong confidence in the expansion of chip manufacturing within the United States. ([reuters.com](https://www.reuters.com/business/media-telecom/softbanks-growing-bets-ai-semiconductor-assets-2025-08-19/?utm_source=openai)) In […]
SoftBank has recently intensified its investments in artificial intelligence (AI) and semiconductor assets, focusing on enhancing U.S. and global technological infrastructure. Notably, the company has committed $2 billion in equity to Intel, signaling strong confidence in the expansion of chip manufacturing within the United States. (reuters.com)
In collaboration with OpenAI and Oracle, SoftBank is leading the Stargate project, a $500 billion data center initiative. This ambitious project aims to bolster AI capabilities and data processing infrastructure. However, the initiative has encountered delays due to prolonged negotiations. Foxconn is slated to manufacture data center equipment for this project in Ohio. (reuters.com)
SoftBank is also at the forefront of a $40 billion funding round for OpenAI, with plans to contribute $22.5 billion by the end of 2025. Additionally, the company supports Perplexity AI, an AI-powered search startup currently valued at $14 billion. (reuters.com)
In March 2025, SoftBank announced the acquisition of Ampere Computing, a chip designer utilizing Arm’s architecture, for $6.5 billion. The company also discreetly acquired British AI chipmaker Graphcore in 2024 and maintains control over Arm Holdings, which went public in 2023 with a valuation of $54.5 billion. Furthermore, SoftBank is rebuilding its stake in Nvidia, holding shares worth $4.8 billion as of June 2025. (reuters.com)
These strategic moves underscore SoftBank’s aggressive expansion across the AI and semiconductor landscape, reflecting its commitment to advancing technological infrastructure and innovation.
John O'Connor is the founder and principal engineer of Web Lifter, a Brisbane software studio building custom software, AI systems, and structured data for Australian SMBs. He has spent over eight years shipping production AI and backend systems, and writes about what actually holds up once the demos are over. Everything published here is drawn from systems running in production for real clients.