Key Takeaways
- • SoftBank injected over $450 million into Graphcore through a single share issuance, representing a significant capital commitment to the British chipmaker less than two years after its 2024 acquisition.
- • The funding signals SoftBank's strategic prioritization of Graphcore as a cornerstone of its semiconductor ambitions, with sources indicating this $450M represents only a portion of expected funding from SoftBank in 2026.
- • Graphcore is accelerating global expansion, including a planned £1 billion AI campus in Bengaluru, as the company rebuilds its commercial presence in the specialized AI chip market.
- • The British chipmaker, founded in 2016 by Nigel Toon and Simon Knowles, was once valued at approximately $2.8 billion in 2020 and positioned as a UK challenger to Nvidia, though it struggled to compete commercially despite strong technical credentials.
- • The investment reflects broader market tailwinds in specialized AI chip development, with the sector projected to grow 30% annually and comprise more than half of the global semiconductor market.
Summary
SoftBank has deployed over $450 million into Graphcore, the British semiconductor company it acquired in 2024, according to filings with Companies House and reporting from CNBC. The capital injection took the form of a single share valued at approximately $457 million, underscoring SoftBank's commitment to revitalizing the chipmaker's business following years of commercial underperformance.
The timing of this funding is strategically significant. Graphcore was once hailed as one of Europe's most credible alternatives to Nvidia during the artificial intelligence boom, particularly after reaching a reported valuation of $2.8 billion in 2020 following funding rounds from Sequoia Capital and Microsoft. However, despite strong technical expertise in intelligence processing units (IPUs)—specialized processors designed for machine learning and complex AI workloads—the company failed to achieve commercial scale as Nvidia consolidated its dominant position in AI infrastructure hardware.
SoftBank's sustained investment signals confidence in Graphcore's long-term potential within its broader semiconductor strategy. A person familiar with the arrangement told CNBC that the $450 million injection represents only a partial commitment, with additional funding expected from SoftBank during 2026. This incremental approach to capital deployment suggests a calibrated strategy for rebuilding the company's commercial operations and market position.
Graphcore is leveraging this capital for aggressive global expansion. The company is planning a £1 billion AI campus in Bengaluru, India, positioning itself to access talent, manufacturing infrastructure, and emerging markets for AI chip deployment. This geographic diversification reflects both the competitive pressures in Western markets and the strategic importance of Asia-Pacific regions in semiconductor and AI infrastructure development.
The broader context reinforces the investment thesis: the specialized AI chip market is projected to expand at 30% annually, with these components expected to comprise more than half of total global semiconductor demand. For SoftBank, which is simultaneously building an AI and robotics spinoff with proposed stakes in OpenAI and recent acquisitions of Ampere Computing, Graphcore represents a diversified play on proprietary AI silicon—a critical component of infrastructure-scale artificial intelligence systems.
The timing of this funding is strategically significant. Graphcore was once hailed as one of Europe's most credible alternatives to Nvidia during the artificial intelligence boom, particularly after reaching a reported valuation of $2.8 billion in 2020 following funding rounds from Sequoia Capital and Microsoft. However, despite strong technical expertise in intelligence processing units (IPUs)—specialized processors designed for machine learning and complex AI workloads—the company failed to achieve commercial scale as Nvidia consolidated its dominant position in AI infrastructure hardware.
SoftBank's sustained investment signals confidence in Graphcore's long-term potential within its broader semiconductor strategy. A person familiar with the arrangement told CNBC that the $450 million injection represents only a partial commitment, with additional funding expected from SoftBank during 2026. This incremental approach to capital deployment suggests a calibrated strategy for rebuilding the company's commercial operations and market position.
Graphcore is leveraging this capital for aggressive global expansion. The company is planning a £1 billion AI campus in Bengaluru, India, positioning itself to access talent, manufacturing infrastructure, and emerging markets for AI chip deployment. This geographic diversification reflects both the competitive pressures in Western markets and the strategic importance of Asia-Pacific regions in semiconductor and AI infrastructure development.
The broader context reinforces the investment thesis: the specialized AI chip market is projected to expand at 30% annually, with these components expected to comprise more than half of total global semiconductor demand. For SoftBank, which is simultaneously building an AI and robotics spinoff with proposed stakes in OpenAI and recent acquisitions of Ampere Computing, Graphcore represents a diversified play on proprietary AI silicon—a critical component of infrastructure-scale artificial intelligence systems.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.