SpaceX buying Cursor exemplifies a new dynamic: entities with massive compute (Colossus) acquire AI applications to turn their capacity into revenue. The application's gross margin problem…

Elon's compute advantage (Colossus) created a unique buyer logic: Cursor's inference cost problem becomes SpaceX's revenue opportunity. This model — compute-rich entities buying application…

In 10x/year markets, acquirers (SpaceX, Stripe) pay strategic value not DCF — Cursor at 10x forward revenue, OpenRouter at 70x trailing; private companies use high-multiple stock as currenc…

Rippling and others cap AI spend at $6-8M/year, selecting 2 models (e.g., Opus 4 + GPT-5.5) not 10; multimodel routing (OpenRouter) serves developers/chatbots but not high-reasoning workflo…

SpaceX ($60B Cursor) and Stripe ($7B OpenRouter + potential $40-50B PayPal) demonstrate private firms can do mega-deals using high-flying stock as currency, closing in weeks without antitru…

Stripe ($7B OpenRouter) and SpaceX ($60B Cursor) demonstrate that well-capitalized private companies can issue stock for transformative acquisitions — bypassing public market scrutiny and t…

Anthropic and Cursor show that while revenue accelerates 10x+ annually, traditional financial constraints (gross margin negativity, massive SBC, compute purchase commitments) are ignored by…

At 12x revenue growth with 40% gross margins, Anthropic's profit is mathematically inevitable; massive SBC (early hires making 50x paper gains) and off-balance-sheet compute commitments are…

In hypergrowth AI, revenue trajectory is the only metric that matters: gross margin negativity, massive SBC paper gains, and compute purchase commitments are all ignored by acquirers and pu…

Top companies are capping AI spend at ~$100K per engineer (running 10 agents 24/7) while cutting dev teams 30-40%. This implies a ~$200B US TAM for coding alone. Rippling's two-model strate…

Workday's closed architecture (hard to build agents on top, hard to churn) justifies Silver Lake's 5.3x revenue LBO price. Open platforms (Salesforce, Monday, HubSpot) face agentic abstract…

Top AI companies now raise on single month's traction (vs 3-4 quarters); meanwhile PE (Silver Lake) targets 20% IRR via precise financial engineering on predictable assets (Workday 5.3x rev…

Workday's closed architecture (hard to build agents on top) justifies a 5.3x revenue take-private multiple because it protects the revenue base from AI disruption; open platforms like Sales…

Workday's closed architecture (hard to build on) protects ecosystem budget capture vs Salesforce's openness which enables agent abstraction; system-of-record moat ensures retention but not…

Cursor, Lovable, Replit, Cognition show that in AI coding, moats build over time through relentless feature velocity (multimodel, security, workflow integration). The winners execute faster…

Cursor, Lovable, Replit have evolved from toys to production-grade platforms with thickening moats via feature depth (security, workflows) and talent density; enterprises standardizing on $…

Products like Cursor, Lovable, and Replit started as simple wrappers but built defensibility by rapidly adding enterprise-grade features (security, workflow integration, multi-model routing…

DOJ's revival of 1914 Clayton Act Section 8 (no interlocking directorates at competitors) targets a16z's Databricks/Fivetran board seats. Initiated under Biden, continuing under Trump. Reso…

The DOJ's use of 1914 Clayton Act Section 8 against a16z for overlapping boards (Databricks/Fiverr) is a low-consequence regulatory action that will likely resolve via board resignations, b…