Good Morning! Welcome to today's AI Business Weekly. You're part of a community that values clarity over hype and business impact over benchmark theater. Let's get to it.
Nvidia Invests Another $2 Billion in CoreWeave as AI Cloud Provider Commits to Multiple Generations of Chips Including Upcoming Rubin GPUs
The investment more than doubled Nvidia's stake in CoreWeave to 11.5% and sent shares surging 15% to $107—the GPU giant is actively cultivating an alternative cloud ecosystem to counter Amazon and Google's push toward proprietary AI chips, while CoreWeave's stock has swung wildly from $183 highs to $64 lows since its March IPO at $40. Read more
Microsoft Loses $381 Billion in Two Days as Wall Street Questions Whether AI Spending Binge Can Ever Pay Off
Shares fell 10% after Azure cloud growth decelerated from 40% to 39% despite quarterly revenue hitting $81.3 billion—the selloff exposed OpenAI's outsized concentration risk, with $281 billion of Microsoft's $625 billion revenue backlog tied to a single customer that doesn't yet turn a profit. Read more
Moltbook Explodes to 150,000 AI Agents in Days as the First Social Network Built Entirely for Bots Raises Security Alarms
OpenClaw agents are posting, debating philosophy, and self-organizing on the platform while cybersecurity firm Palo Alto Networks warned of a "lethal trifecta" of vulnerabilities—a critical database exploit allowed anyone to hijack any agent on the platform before it was temporarily taken offline, and X's own AI assistant Grok is actively mislabeling verified videos as AI-generated. Read more
British Columbia Forces AI and Data Centre Companies to Compete for Electricity as Grid Strain Hits North America
BC Hydro opened a 400-megawatt competitive bid under Bill 31, explicitly citing rate hikes in New Jersey and Virginia as a cautionary tale—the province is allocating roughly 35% of its Site C dam output to AI and data centre projects while Bell Canada has already announced six facilities and former bitcoin miners are converting to AI workloads. Read more
AI-Manipulated Images of Minneapolis Shooting Spread Across Social Media, Fooling Senators and Millions of Users
A U.S. senator displayed an AI-altered image on the Senate floor without realizing it was fabricated, while a deepfake video on Facebook depicting a fabricated shooting scenario accumulated over 44 million views—experts warn the episode marks a new phase where AI enhancements of real photos are indistinguishable enough to mislead lawmakers while simultaneously eroding trust in verified media. Read more
📢 The Signal Behind the Noise
This edition spans five very different stories that share a single thread: the gap between AI's promises and the infrastructure, security, and trust systems required to deliver on them. Nvidia doubling down on CoreWeave signals the GPU ecosystem is diversifying, but Microsoft's $381 billion selloff proves Wall Street has stopped rewarding AI spending on faith alone—it now demands proof of returns. Moltbook's 150,000-agent experiment is a genuine stress test of autonomous AI at scale, and it failed its first security audit within days. British Columbia is the first major jurisdiction to formally ration clean power among AI projects, because the grid simply cannot keep pace. And the Minneapolis deepfakes don't just blur one event—they demonstrate that AI-manipulated media has crossed the threshold from niche concern to active threat to public discourse. The technology is outrunning every system designed to govern it.
