Chip Startup

• Situational Awareness, an AI-focused hedge fund founded by former OpenAI researcher Leopold Aschenbrenner, has invested $400 million in chip startup Source Foundry. • This latest funding round brings the fund's total investment in the Stanford-founded startup to $500 million. • Source Foundry aims to revolutionize the semiconductor industry by making the chip manufacturing process both faster and more affordable.

• Hedge fund Situational Awareness has reportedly invested $400 million to $500 million in Source Foundry, a semiconductor startup valued at $5 billion. • Source Foundry aims to reinvent chip manufacturing by developing advanced lithography tools to address critical bottlenecks in the AI supply chain. • This aggressive bet comes shortly after Situational Awareness experienced what has been described as the "most catastrophic hedge fund blowup" of the year.

• Hedge fund Situational Awareness has committed a total of $500 million to Source Foundry, including a fresh $400 million injection this week. • Founded last year by Stanford researchers Abdulmalik Obaid and Joe Burg, Source Foundry develops specialized equipment to resolve bottlenecks in advanced semiconductor manufacturing. • The investment follows a recent funding round that valued the startup at $5 billion, highlighting the high demand for domestic chip production efficiency.

• South Korean venture capital firms are aggressively investing in local semiconductor material startups to capitalize on the escalating global AI chip race. • This surge in funding is driven by the intensifying competition for dominance in artificial intelligence hardware and the critical need for advanced materials. • The trend highlights South Korea's strategic effort to strengthen its domestic supply chain and reduce reliance on foreign semiconductor components.

• Chinese regulators are reportedly tightening controls on the export of artificial intelligence technology following the cancellation of Meta's $2 billion acquisition of AI firm Manus. • Domestic AI startups and semiconductor companies have been warned to prevent cutting-edge technologies, critical training data, and top technical talent from being acquired by Western entities. • This move signals a strategic shift by Beijing to protect national security and maintain a competitive edge in the global AI race by restricting the outflow of intellectual property.

• Jeff Bezos and AMD's venture unit have invested in CuspAI, an AI startup focused on discovering new materials for chipmaking. • Originally centered on carbon capture and water purification, the company is expanding its scope to accelerate the development of semiconductor materials. • To support this growth, CuspAI hired former Google and Apple executive John Giannandrea to help establish a new operational outpost in California.

• Jeff Bezos and firms Kleiner Perkins and NEA have invested in CuspAI, a British startup focused on using AI to solve physical-world challenges. • The company raised $450 million in a funding round that values the startup at $2.6 billion. • CuspAI has partnered with Nvidia and other industry leaders to accelerate the discovery of materials for semiconductors, clean energy, and advanced manufacturing.

• Chinese AI startup DeepSeek is developing its own proprietary AI chips to power its artificial intelligence systems, according to a report by Reuters. • The move comes as the company seeks to reduce reliance on external hardware providers to sustain its computational needs. • This development is significant as China faces ongoing U.S. export restrictions on high-end semiconductors, such as those from Nvidia.

• French AI startup ZML, backed by Turing Award winner Yann LeCun, has released ZML/LLMD, a free software tool designed to accelerate AI inference across various chips. • The software aims to reduce the operational costs of running AI models, positioning itself as a competitor to existing tools like vLLM and SGLang. • Led by a lean, well-funded team of 20 people in Paris, ZML is expanding its ambitions toward the co-design of silicon hardware.

Amazon Web Services began deploying AI chips from Groq alongside its Trainium2 processors in US data centers on March 13, 2026, to diversify inference capabilities. The move supports over 100,000 Inferentia chips already in use, aiming to cut AI training costs by 50% for customers. This hybrid strategy enhances AWS competitiveness against Nvidia amid chip shortages, boosting scalability for enterprise AI workloads. Expansion to additional regions is slated for April 2026.