The market doesn't pause to let you catch up. Here are the five stories worth your attention this week — and what they actually mean if you're building, investing, or trying to stay a step ahead.
Kleiner Perkins raises $3.5B, all of it pointed at AI
KP closed $3.5 billion across two new funds: $1B for early-stage (KP22) and $2.5B targeting growth-stage companies. That brings the firm’s total AUM past $21 billion. For context, the previous flagship raise was just over $2 billion, less than two years ago.
Spencer’s take: When firms like KP 75x their raise cadence, it’s not FOMO — it’s conviction backed by exits. Figma IPO’d. Windsurf sold to Google. The scoreboard is moving. Expect valuations on AI infrastructure plays to compress the window for seed-stage entry even further.
Panthalassa raises $140M to build wave-powered AI data centers
Panthalassa, a Pacific Northwest climate-tech startup, closed a $140 million Series B led by Peter Thiel. The Oregon company is building floating data-center platforms powered by wave energy — giant ocean buoys paired with onboard AI computing, with data transmitted via low-Earth-orbit satellites. Commercial deployments are planned by 2027.
Spencer’s take: Peter Thiel funding floating ocean servers is either the most Peter Thiel thing that’s ever happened, or it’s a preview of where the compute arms race actually goes. Energy-constrained AI creates off-grid compute arbitrage. Don’t sleep on this.
Reserv closes $125M Series C to automate insurance claims
Reserv, a New York-London insurtech, closed a $125 million Series C led by KKR. The company’s platform uses generative AI to automate claims processing for insurers and third-party administrators — and claims it can double or triple an insurer’s claims-handling capacity while reducing reliance on human adjusters.
Spencer’s take: Insurance claims are the back-office problem no one glamorizes, which is exactly why it’s a great business. Boring workflows with huge volume and zero incumbents who want to cannibalize themselves. KKR doesn’t do charity, they see the margin.
Moonshot AI raises $2B at a $20B valuation
Chinese AI startup Moonshot AI announced a roughly $2 billion funding round valuing the company at about $20 billion. Known for its “Kimi” series of open-weight large language models, the round was led by Meituan’s VC arm with participation from Tsinghua Holdings and China Mobile. In the past six months alone, Moonshot raised nearly $3.9 billion, more than doubling its valuation since late 2025.
Spencer’s take: The US-China AI race isn’t theoretical anymore. Moonshot’s K2.6 model is one of the most-used LLMs in China right now. If you think the competition is just OpenAI vs. Anthropic, you’re watching the wrong game.
AI captured 41% of all venture dollars on Carta in 2025
Carta data confirmed that AI startups captured 41% of all venture dollars on the platform in 2025, a record high. That’s not a sector dominating that’s a sector absorbing capital that used to go everywhere else.
Spencer’s take: Every other category is now competing for the remaining 59 cents of every VC dollar. Fintech, biotech, climate they’re all fighting for scraps unless they can attach an AI narrative. That creates real opportunity for founders who build in underinvested spaces. It also means the next non-AI unicorn will be dramatically underfunded on the way up.
The #1 Story: KP’s $3.5B and the Concentration Effect
The Kleiner story is the most important one this week, and it’s best understood through the mental model of Second-Order Thinking.
First-order: “More money = more AI startups = more competition.” That’s the obvious read. But the second-order effect is more interesting. When the biggest funds get bigger and more concentrated, deal access becomes the scarcest resource — not capital. Thrive Capital just secured $10 billion, General Catalyst is targeting a comparable amount, and Founders Fund closed $6 billion for its own growth-stage vehicle. The result is an oligopoly of firms competing for the same 20 breakout deals, which pushes seed valuations up and creates a barbell market: richly funded AI infrastructure plays at the top, and a long tail of bootstrapped or barely-funded companies at the bottom. The middle gets hollowed out.
If you’re building at the intersection of regulated markets and emerging tech — like we are at /mkt with athlete tokenization under a Reg A+ structure — that barbell is actually an opening. Compliance complexity keeps big generalist funds away. That’s a moat, not a barrier.
Next Week on Startup Spotlight
I’m going deep on one company that’s quietly building the financial infrastructure layer no one’s talking about. It’s not another AI wrapper. It’s a real business with real compliance requirements and a thesis that could age very well. Paid subscribers get the full breakdown.