Good morning.
Today's briefing unpacks the immense capital and strategic maneuvering shaping the future of artificial intelligence. We'll explore OpenAI's unprecedented, trillion-dollar gamble on physical infrastructure, a move designed to secure dominance in the next wave of AI development. This theme of tangible technology continues with SoftBank's major acquisition in the industrial robotics sector, signaling a clear bet on "physical AI." Meanwhile, the practical application of AI is rapidly transforming core business functions, as evidenced by Zendesk's new autonomous agents set to overhaul the customer service industry.
Physical AI. SoftBank's $5.375 billion acquisition of ABB's robotics unit signals a major strategic pivot towards tangible, real-world AI applications. Chairman Masayoshi Son articulated a vision to "fuse Artificial Super Intelligence and robotics," positioning this move as the company's "next frontier." This deal integrates a major industrial robotics manufacturer into SoftBank's existing portfolio of AI-focused investments, aiming to drive a groundbreaking evolution and reignite growth in the robotics sector. It underscores a significant long-term bet that the future of AI lies not just in software but in intelligent, autonomous physical systems.
Autonomous Workforce. Customer service giant Zendesk is launching a suite of LLM-driven autonomous agents capable of resolving 80% of support issues without human intervention. This move represents a fundamental shift in the industry, which Zendesk's product president described as moving "to a system where AI actually does most of the work." The strategy, bolstered by recent acquisitions, has already shown a five to 10-point increase in customer satisfaction during previews. This development signals a major acceleration in workforce automation with profound implications for the 2.4 million customer service representatives in the U.S. alone.
Global Growth. OpenAI is aggressively expanding its affordable ChatGPT Go subscription plan to 16 additional countries across Asia, a strategic play to capture high-growth emerging markets. The company reports a fourfold increase in its weekly active user base in Southeast Asia since the plan's initial rollout, demonstrating strong regional demand for accessible AI tools. This expansion intensifies competition with Google and is a key part of OpenAI's strategy to achieve profitability while expanding its global user base, which now stands at 800 million weekly active users.
Retail Innovation. Google is enhancing its e-commerce capabilities by expanding its AI-powered virtual try-on feature internationally and introducing functionality for shoes. This technology allows consumers to see products on a virtual version of their own body, moving beyond generic models to create a more personalized and engaging online shopping experience. The move represents a strategic shift towards generative AI in retail, placing Google in direct competition with similar offerings from Amazon and Walmart, with the goal of reducing return rates and increasing conversion.
Deep Dive
The race to build the next generation of artificial intelligence is no longer just about algorithms; it's a colossal battle for physical infrastructure. OpenAI is at the forefront of this shift, moving from a research lab to a global-scale utility provider. CEO Sam Altman's recent confirmation that "much more" is coming in terms of infrastructure deals signals a deliberate, high-stakes strategy to secure the massive computational power required for future AI models, a bet that the demand for advanced AI will be so profound that it justifies an unprecedented capital outlay.
The scale of this bet is staggering. In 2025 alone, OpenAI has reportedly inked nearly $1 trillion in deals, including a $500 billion "Stargate" project with Oracle and SoftBank to commission ten gigawatts of U.S. facilities. This is coupled with a $100 billion investment from Nvidia, which is now actively helping OpenAI become its own "self-hosted hyperscaler." Nvidia's CEO Jensen Huang estimates each gigawatt of AI data center infrastructure could cost between $50 billion and $60 billion, putting the sheer magnitude of OpenAI's financial commitments into perspective.
This "very aggressive infrastructure bet," as Altman calls it, has profound strategic implications. It aims to create an insurmountable competitive moat built on raw compute power, making it exceedingly difficult for rivals to catch up. However, it also introduces immense financial risk, with expenditures far outpacing current revenues of $4.5 billion in the first half of 2025. This strategy transforms the AI landscape into a capital-intensive game playable only by a few, raising questions about industry consolidation and the long-term sustainability of a model that bets the house on future, yet-unrealized, technological breakthroughs.
