HUMAI is your guide to the intersection of artificial intelligence, digital tools, and the future of human creativity. Explore curated insights, experiments, and high-impact resources for creators, strategists, and builders.
Anthropic committed $200 billion to Google Cloud over five years, locking in 5 gigawatts of TPU capacity. Year-one bills are larger than Anthropic's current revenue. Alphabet now sits within $400 billion of overtaking Nvidia as the world's largest company.
Microsoft posted record Q3 profit of $31.8B and 40% Azure growth. In the same week it announced a $900M charge to push 8,750 senior employees out via the first voluntary buyout in its 51-year history. The math behind the contradiction.
Alphabet's $62.6B Q1 profit jumped 81%. Nearly half came from writing up its Anthropic stake. Amazon booked $16.8B from the same source. Both companies set the price they marked themselves to.
Apple set a record March quarter at $111 billion and another $100 billion buyback. Then Tim Cook warned memory costs will hit hard, admitted Siri's brain is rented from Google, and hinted at a major AI acquisition.
Google Cloud's operating margin went from 9.4% to 32.9% in twelve months. Cloud grew 63%. Wall Street rewarded Alphabet with $200B in market cap and punished Microsoft, Meta and Amazon for spending the same money. The split is structural.
OpenAI missed its 1B weekly user target. Monthly revenue goals slipped. CFO Sarah Friar told the board the company isn't IPO-ready. Compute commitments now total roughly $1 trillion against $25B of annualized revenue. Big Tech earnings tonight settle the next move.
China's NDRC ordered Meta to unwind a $2 billion Manus acquisition that closed five months ago. Engineers are already badged in. Money already moved. Beijing just claimed jurisdiction over a Singapore company. Every Chinese-founded AI startup is now a hostage.
Tesla beat earnings, but strip out carbon credits and Bitcoin gains and the core auto business cleared $21 million on $19 billion in vehicle revenue. Free cash flow turns negative for the rest of 2026. The Model S line dies in May. Optimus has to work.
DeepSeek released V4-Pro on Friday: open-weight, near-frontier on coding and reasoning, and priced at one-seventh of GPT-5.5. The model layer just became a commodity. OpenAI and Anthropic now have to defend $1.6 trillion in combined paper value with a moat that no longer exists.
ServiceNow delivered a clean Q1 beat and raised its 2026 AI revenue forecast by 50 percent. The next morning, the stock had its worst day on record. The market is now repricing every per-seat SaaS company at once.
Meta laid off 10 percent of its workforce the week before earnings. Zuckerberg is spending $135 billion on AI in 2026. Llama still ranks last among the major US labs. Here is the pressure math.
Anthropic refused to let Claude be used in autonomous weapons or mass surveillance. The Pentagon blacklisted it, OpenAI took the $200M contract, and the CFO says billions in 2026 revenue are on the line.
Amazon put $50 billion into OpenAI in February and $25 billion into Anthropic on Monday. Both labs agreed to spend $100 billion back on AWS compute. This is circular AI finance in its cleanest form, and it only works if revenue keeps compounding at 1,400 percent.
Maine passed the first state-wide ban on large data centers on April 13. At least 25 US projects were already killed in 2025 by local opposition. The hyperscalers still plan to spend $660 billion on AI infra in 2026. The capex math is breaking.
Oracle laid off 18 percent of its workforce in March to pay for AI data centers. The $553 billion backlog looks huge until you notice over half of it is one customer, OpenAI, and the revenue does not start until 2027.
Jack Dorsey cut 40 percent of Block's workforce in February, citing AI productivity gains. Weeks later, a senior engineer threatened to quit unless his team was rehired. The gap between the public story and the private corrections is now the real pressure.
OpenAI's $852 billion valuation faces growing skepticism from its own backers. With $14 billion in projected losses, Anthropic overtaking revenue, and an internal fight over IPO timing, the pressure points are becoming visible.
Microsoft just launched three home-built AI models that directly compete with OpenAI's most commercial products. Given that Microsoft has invested $13 billion in OpenAI, the implications are hard to ignore.
