20VC: OpenAI and Anthropic Threatened by Kimi? | Should the US Ban Chinese Open-Source Models | Should Openrouter Sell & Value in the Routing Layer? | Stripe Buying Paypal: What You Need to Know

What It Covers

Harry Stebbings, Rory O'Driscoll, and Jason Lemkin analyze five major tech developments: China's Kimi and Qwen open-weight model releases, Washington's response to Chinese AI, Databricks' $3B Series M at $188B valuation, OpenRouter's reported acquisition talks, Fireworks AI's $1.5B round, and Stripe's reported bid to acquire PayPal.

Key Questions Answered

• •Chinese Open-Weight Model Threat: Kimi K3 is a 2.8 trillion parameter model comparable in size to US frontier models, not a lightweight alternative. On OpenRouter, 50% of traffic already runs through Chinese-created models. The real risk isn't a sudden takeover — it's gradual enterprise adoption accelerating as cost differentials reach 80% cheaper than closed frontier models like Claude or GPT-4.
• •OpenRouter Sale Timing: The optimal window to sell a routing/aggregation layer business is precisely when commodification begins but before acquirers fully recognize it. OpenRouter's last round valued it near $1.8B; a $5–6B exit represents the crossover moment where hyperscalers — particularly Amazon, which positions itself as model-agnostic — would pay a strategic premium far above standalone NPV to shift enterprise market share.
• •Founder Sale Decision Framework: A 3x return on last round is insufficient justification for a founder to sell. The threshold should be 10x, because the risk-adjusted math changes entirely at scale. A founder holding 12–15% of a $1.8B company walking away with $600–700M at a $5–6B exit faces enormous execution risk for a relatively marginal personal wealth increase versus taking the certain outcome.
• •Infrastructure vs. Application Layer Gap: AI infrastructure spending runs at $800–900B annually. The two frontier model companies generate roughly $100B combined. Every other AI application company combined — including Cursor at $4B ARR — totals under $40–50B. The application layer renaissance has not arrived; spending on training data for foundation models alone likely exceeds the sum of all non-Cursor AI application revenues.
• •Foundation Model Growth Rate as Market Bellwether: The single variable determining the trajectory of US tech valuations, hyperscaler RPO commitments, and AI infrastructure CapEx is OpenAI and Anthropic's revenue growth rate in 2026–2027. If either company slips from 10x annual growth to 2–3x, commitments made against that growth assumption trigger a broad market dislocation across every adjacent sector simultaneously.
• •Stripe-PayPal Acquisition Dance: Stripe's reported offer carries a 28% premium to PayPal's public stock price; historical take-private averages land in the mid-30s. The board's rejection is a procedural negotiating move, not a genuine refusal. Delaware fiduciary duty requires PayPal's board to demonstrate a credible standalone plan exceeding the offer's value — difficult given three consecutive failed CEO turnaround attempts and flat internal metrics.