The race to build AI infrastructure was constrained not by advanced semiconductors, but by the availability of power transformers. This overlooked, 100-year-old technology saw lead times balloon to over three years, becoming the single biggest gating factor for new data center deployments.
A key, under-the-radar outcome of the Epic vs. Google lawsuit is the creation of a new app class: 'metaverse browsers.' These apps will allow users to navigate virtual worlds with portable digital items, signaling a concrete step toward building the foundational layer for an interoperable metaverse.
While the unit cost of AI inference has plummeted 50x, overall spending on AI is surging. This is a textbook example of Jevons paradox, where radical efficiency gains lead to increased consumption and higher total expenditure as new applications become economically viable.
The future of hardware testing involves moving beyond simple, sequential pass/fail checks. AI test agents will instead explore a system's state space, intelligently choosing the next test point that will yield the most new information, a concept called 'knowledge maximizing.'
AI is automating the task of writing code, leading to a decline in "programming" jobs. Simultaneously, demand for "software engineering" roles, which involve higher-level system design and managing AI tools, is growing. This signals a fundamental reskilling shift from pure coding to architectural oversight.
Two years into the AGI boom, the vast majority of market value accrued to infrastructure providers like NVIDIA ($3.2T gain). In contrast, major platform players like Microsoft saw minimal gains (4%), proving the "picks and shovels" strategy was the definitive winner.
Mutual fund giants like Fidelity invest in late-stage startups less for the potential return and more to build relationships that guarantee them a significant allocation when the company goes public. This access is a key value proposition they offer to their own high-net-worth and institutional clients.
AI's impact on inequality is dual-faceted. It may reduce the wage gap by automating high-skill jobs faster than low-skill ones. However, it simultaneously increases wealth inequality by concentrating massive capital gains within a few dominant tech companies, benefiting asset owners disproportionately.
Companies like Stripe are avoiding IPOs because the private markets now solve the two main historical drivers: access to capital and employee liquidity. With annual secondary tenders and vast private funding available, the traditional benefits of going public are no longer compelling for many late-stage startups.
Daniel Gross's prescient question about copper being mispriced proved correct. The metal hit all-time highs due to AI's physical needs, with a single NVIDIA server rack containing two miles of copper wire. This highlights a critical, non-obvious bottleneck in the AI supply chain.
SpaceX is targeting a monumental $1.75T IPO valuation that cannot be justified by its current financials. The strategy relies on Elon Musk's powerful narrative-building and his history of achieving seemingly impossible goals, framing the IPO as a controlled liquidity event rather than a price discovery based on fundamentals.
Despite median venture capital funds lagging public indexes like the S&P 500 for a quarter-century, capital continues to pour into the asset class. One LP describes this as 'hope over experience,' as investors are lured by the outlier returns of top funds, even though the average dollar invested underperforms.
Contrary to the theory that a nation could achieve AGI by using vast amounts of cheap energy to power older chips, evidence shows this is not viable. All frontier models to date have been trained on the most advanced semiconductor nodes (5nm or less), indicating that architectural efficiency is a non-negotiable requirement.
Apple's inability to ship its own cutting-edge AI model has paradoxically become a strategic advantage. Instead of bearing the immense cost of foundation model development, they can now integrate best-in-class third-party models onto their dominant hardware ecosystem, a position Mark Gurman calls 'falling ass backwards into it.'