The Infrastructure Gold Rush Continues

The first quarter of 2026 provided a stark, two-sided picture of the AI economy: massive returns and even more massive future spending. Big Tech just proved AI infrastructure spending works. Then it raised the bill anyway. Microsoft, Alphabet, Meta, and Amazon collectively committed somewhere between $630 billion and $650 billion in capital expenditure forecasts, signaling that the race for compute is far from over. The results justified the spend: Google Cloud surpasses $20B, but says growth was capacity-constrained, and Amazon’s cloud business is surging — and so is its capital spending. Microsoft CEO Satya Nadella, armed with a new deal that lets Microsoft offer OpenAI’s tech without paying for it, declared he’s ready to ‘exploit’ the new OpenAI deal. On the product side, Microsoft says it has over 20M paid Copilot users, and they really are using it, challenging the perception of low adoption.

Startup Valuation Soars, User Sentiment Sinks

While the infrastructure giants plow ahead, a leading AI startup is reportedly nearing stratospheric valuation. Sources: Anthropic could raise a new $50B round at a valuation of $900B, a figure that underscores the perceived value of frontier AI models. However, a contrasting trend emerges in user sentiment. A new study suggests The more young people use AI, the more they hate it, indicating that early and heavy adoption among Gen Z may be leading to disillusionment with the current capabilities and applications of LLMs.

Governance Gaps and Regulatory Warnings

As AI integration deepens, regulatory scrutiny intensifies. AI agent governance takes focus as regulators flag control gaps. Australia’s financial regulator warned that AI agent governance and assurance practices are “poorly governed” at financial firms, highlighting a growing concern as autonomous systems handle more critical tasks. This technical oversight is paralleled by a need for clearer technical foundations. A primer aims to demystify APIs, MCPs, and MCP Gateways, explaining the crucial differences between these methods of system communication.

Meta’s AI Push Amid Platform Decline

Meta presented a mixed financial picture. While Meta says its business AI now facilitates 10 million conversations a week and over 8 billion advertisers have used its gen AI tools, the company also revealed Meta lost 20 million users last quarter. Despite this user decline, Meta plans to pump billions more into AI investments this year and is still burning money on AR/VR through its Reality Labs division.

Security, Spoofs, and Strange Model Quirks

OpenAI is preparing a specialized, limited-release cybersecurity model. OpenAI’s new security model is for ‘critical cyber defenders’ only, GPT-5.5-Cyber, will not be available to the public initially. The company also addressed a peculiar model behavior, publishing an explanation Where the goblins came from after reports that its coding model was instructed to avoid talking about certain creatures. In the creative world, Spotify launched a Verified by Spotify badge lets you know this artist isn’t AI to combat spam and AI-generated impersonations.

Robotics and Legal Drama

In robotics, SoftBank is creating a robotics company that builds data centers — and already eyeing a $100B IPO, a move that embodies the cyclical nature of AI infrastructure. In the courtroom, the legal battle between Elon Musk and OpenAI continued, with reports noting that On the stand, Elon Musk can’t escape his own tweets and Elon Musk’s worst enemy in court is Elon Musk, as his own past statements and testimony complicated his case.

Editorial: The Efficiency Paradox

Today’s news highlights a central paradox of the current AI boom: the technology is generating unprecedented revenue and utility, yet the cost to develop and deploy it is scaling at an even faster rate. The cloud profits are real, but they are immediately reinvested into an even larger bill for next quarter. This suggests we are not in a phase of optimization, but of raw expansion. Meanwhile, as the tools become more powerful and integrated (10 million business conversations a week), the calls for governance and the signs of user fatigue grow louder. The industry is building at a breakneck pace, but the foundational questions of control, cost, and human reception remain unresolved.