Cursor's $60B deal

Good morning,

AI is moving beyond software into the physical world, while capital and control battles intensify. From robots beating humans to $60B acquisitions and model leaks, the stakes are rising fast.

Let’s dive in 👇

🧠 AI Breakthroughs & Power Plays

🏓 Robots Beat Humans… For Real

Ping pong robot ace makes history as an AI-powered system defeated elite human players in competitive table tennis, marking a real milestone in physical AI. This wasn’t simulation or lab conditions, it handled real-time speed, spin, and unpredictability. It signals that embodied AI, not just chatbots, is advancing fast.

🚗 Tesla Doubles Down on AI

Tesla’s $25 billion spending plan is testing investor patience as the company pours massive capital into AI initiatives with unclear timelines. The focus spans autonomy, robotics, and infrastructure, but returns remain speculative. This is a high-risk bet that Tesla can translate AI into real revenue.

🔓 Anthropic Model Access Leak

Anthropic’s Mythos model accessed by unauthorized users raises concerns around model security and distribution control. It suggests that even top labs are struggling to fully contain access to frontier systems. This ties directly into the broader issue of “intelligence leakage.”

🏢 Enterprise AI Wars

👀 Meta Tracking Employee AI Usage

Meta tracks employee usage on Google and LinkedIn as part of internal AI training initiatives, signaling how seriously companies are monitoring adoption. This is less about surveillance and more about optimizing how workers use AI tools. Expect more companies to quantify AI productivity internally.

🚀 SpaceX Makes Move on Cursor

SpaceX strikes $60B deal to secure rights tied to Cursor, one of the fastest-growing AI coding platforms. This shows how strategic coding agents have become, not just tools, but infrastructure bets. The valuation reflects how critical developer workflows are becoming.

⚡ Google Fires Back at Nvidia

Google Cloud unveils new TPU chips aimed directly at competing with Nvidia’s dominance in AI compute. The chips are optimized for large-scale model training and inference. This is part of a broader shift where hyperscalers want full-stack control.

🛠️ Tools of the Day

→ Kollab - collaborative AI workspace for teams
→ Monid - AI-powered identity and monitoring layer
→ Claude Code UltraReview - advanced code review and optimization assistant

⚡ Quick Hits

→ Anthropic code leak raises copyright concerns
→ Google Workspace AI pushes AI deeper into daily work
→ OpenAI + Infosys expands enterprise reach
→ AI drug discovery surge creates validation bottleneck

🧾 TLDR

AI is expanding beyond software into the physical world, massive capital bets, and infrastructure wars. Tesla and SpaceX are making aggressive long-term plays, while Google pushes hardware to compete with Nvidia. At the same time, leaks and internal tracking show that control over AI usage and access is becoming a central issue.

Cheers,
David