TL;DR

xAI raises $9.3B; Meta’s $32B SSI bid fails, pivots to talent; Surge AI hits $1B revenue bootstrapped.

Highlights

• xAI raised $9.3B, will spend $4.7B in three months, and targets $18B for data centers by 2027, expanding its AI supercomputer to 340,000 GPUs1.
• Meta Platforms, Inc. Class A Common Stock ’s $32B acquisition bid for Safe Superintelligence (SSI) was rejected; Meta Platforms, Inc. Class A Common Stock is now hiring SSI’s CEO and invested $14.3B for 49% of Scale AI24.
• SoftBank is exploring a $1T AI and robotics manufacturing hub in Arizona (“Project Crystal Land”), seeking Taiwan Semiconductor Manufacturing Company Ltd. as an anchor partner; still in early stages3.
• Surge AI, a bootstrapped data labeling startup, surpassed Scale AI with $1B in 2024 revenue, remaining profitable and VC-free10.
• SK Group and Amazon.Com Inc will invest $5.1B to build South Korea’s largest AI data center, with plans to expand tenfold11.
• Nvidia Corp and Foxconn plan to deploy humanoid robots at a new Houston AI server plant by 20268.
• Tesla, Inc. Common Stock is piloting a commercial robotaxi service in Austin with 10 geofenced vehicles; regulatory scrutiny is ongoing13.
• Archer Aviation Inc. secured a $250M deal for eVTOL deployment in Indonesia, formed a five-country certification alliance, and partnered with Jetex7.
• France will take a 30% stake in Eutelsat as part of a €1.35B capital raise to expand its satellite network6.
• Parataxis Holdings acquired Bridge Biotherapeutics for $18.5M to launch a Bitcoin treasury platform in South Korea9.
• Reddit, Inc. is considering integrating World ID’s iris-scanning biometric tech for user verification15.
• Deezer will flag AI-generated music and withhold royalties from fraudulent streams, pressuring larger streaming platforms14.

Commentary

AI infrastructure and talent acquisition continue to drive the largest capital allocations and strategic moves in the market. xAI’s rapid $4.7B deployment and $18B multi-year capex plan for data centers underline the capital intensity required to compete at the frontier of AI model training and hardware1. Meta Platforms, Inc. Class A Common Stock ’s aggressive, but unsuccessful, $32B bid for SSI—followed by a major talent acquisition and a $14.3B stake in Scale AI—reflects the escalating costs and competition for both talent and data assets24. These developments are setting new benchmarks for deal size and speed in the AI sector, with implications for both early- and growth-stage deal valuations.

SoftBank’s early-stage $1T “Project Crystal Land” proposal signals ambitions to anchor the U.S. as a global center for AI and robotics manufacturing, though execution risk is high given the project’s scale and dependence on partners like Taiwan Semiconductor Manufacturing Company Ltd. 3. Meanwhile, the SK Group/Amazon.Com Inc $5.1B data center in Korea and Nvidia Corp -Foxconn’s humanoid robot plans in Houston highlight the global race to scale compute and automation, with downstream opportunities for startups in chips, robotics, and supporting infrastructure811.

Surge AI’s $1B revenue milestone, achieved without outside funding, is a notable outlier in the data labeling space and suggests that capital-efficient, profitable models can still outperform heavily funded competitors10. This challenges the prevailing VC narrative in AI infrastructure and should prompt investors to re-examine assumptions about market consolidation and capital requirements.

On the application side, Tesla, Inc. Common Stock ’s robotaxi pilot and Archer Aviation Inc. ’s international eVTOL deals demonstrate that regulatory and certification hurdles remain, but commercial pilots are progressing713. The Eutelsat recapitalization and France’s increased stake reinforce the strategic importance of satellite connectivity6, while Parataxis Holdings’ move into Bitcoin treasury management via biotech M&A signals continued crossover between digital assets and traditional sectors9. Reddit, Inc. ’s potential adoption of World ID and Deezer’s AI-content policy reflect the growing focus on digital identity and content integrity1415.