Four disclosed rounds this week added up to roughly $2.34 billion. Three of them funded physical infrastructure, regulated infrastructure, or classified work. Almost none of it was pure software. Here's what mattered, in order.

1. Travis Kalanick’s Atoms raised $1.7 billion, and Uber wrote a check TechCrunch reported on July 22 that Atoms, Kalanick’s industrial robotics holding company, closed $1.7 billion in equity led by Andreessen Horowitz, with Ben Horowitz joining the board. Bain Capital, Fifth Wall, K5 Global, SV Angel and Uber participated. Atoms sits on top of CloudKitchens and Pronto and now runs three divisions: Food, Mining and Transport. The company hasn’t disclosed a post-money valuation, a debt facility size, or commercial traction figures for any of those divisions.

Spence’s hot take: When the company that pushed you out becomes an investor, that isn’t forgiveness. That’s a supply chain decision.

The model: Asymmetric Information Asymmetric information means one side of a transaction knows things the other side doesn’t, and the price gets set by the informed side, not by you. a16z, Bain and Uber saw a data room before they wired $1.7 billion. The rest of us saw a founder’s post on X and a vision letter. Pay attention to what got disclosed here (round size, lead investor, board seat) versus what didn’t (valuation, revenue, units deployed). Companies release the numbers that help them raise the next round. The silence is data too.

For sector context: Crunchbase reported in late June that robotics startups had raised $18.8 billion globally in 2026, already ahead of the $15 billion raised across all of 2025.

2. OpenAI added two bank CEOs to its board CNBC reported that OpenAI appointed Nubank founder and CEO David Vélez and BNY CEO Robin Vince to its nonprofit and for-profit boards on July 21. Vince spent more than 25 years at Goldman Sachs before taking over BNY in 2022, and reporting indicates he’ll chair the audit committee. OpenAI, valued at more than $850 billion according to CNBC, filed its prospectus confidentially with the SEC in June and hasn’t announced a listing timeline.

Spence’s hot take: Nobody recruits an audit committee chair out of Goldman because they’re curious about accounting.

3. Four DOGE alumni raised $160 million for a stealth military cyber startup Reuters reported on July 22, citing three people familiar with the matter, that Cathedral closed $160 million at a $1.4 billion valuation co-led by a16z and Sequoia, with both firms taking board seats. Founders Gavin Kliger, Luke Farritor, Marko Elez and Jack Stein are building AI tooling for offensive and defensive military cyber work. Cathedral declined to comment, and no customers, contracts or revenue figures have been publicly disclosed. Treat that valuation as unconfirmed reporting from unnamed sources, not a company statement.

Spence’s hot take: A $1.4 billion price tag before a public product is a bet on access, not on code.

4. Augustus raised $180 million at a $1 billion valuation to build a “Global Dollar Bank” Per the company’s July 21 release, Tiger Global led the Series B with Hummingbird, QED, and founders from Nubank, Ramp, Circle and Deel participating. Augustus says it has raised $210 million to date and already processes volume for customers including Kraken. The pitch is clearing infrastructure that runs across Swift, ACH, SEPA and stablecoin rails instead of issuing its own token. Worth flagging a discrepancy: the company describes itself as a federally chartered bank, while Decrypt reported the OCC charter is conditional. Those aren’t the same thing.

Spence’s hot take: The winners here won’t be the fastest teams. They’ll be the ones who can survive an exam. That’s the same reason we build on regulated rails at /mkt: slower to stand up, much harder to copy.

5. Sila raised $300 million for silicon anodes while the EV market wobbles Sila announced on July 21 via BusinessWire that Atreides Management and Sutter Hill Ventures led a $300 million private round, joined by 8VC, Bessemer Venture Partners, Matrix Partners and funds advised by T. Rowe Price. The capital funds Phase 2 at the company’s 160-acre Moses Lake, Washington plant. Sila says its Titan Silicon material delivers 20 to 40 percent higher energy density than graphite, and TechCrunch reported the expansion targets enough anode material for more than 100,000 EVs.

Spence’s hot take: Sila quietly stopped leading with the EV story and started selling to drones, satellites, robotics and AI hardware. Same material, better customer.

Next week’s Startup Spotlight deep-dive: the charter race. What actually happens when stablecoin rails collide with federally regulated banking, why the boring compliance work is the moat, and which side of that collision the capital is picking.

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, on Amazon right now.

Next week on Startup Spotlight

Every company above made noise this week. Next Wednesday’s paid deep-dive goes the other direction: the biggest AI infrastructure bet of the moment might be the one that never touches the cloud. I’m breaking down the on-prem AI shift, who’s quietly winning it, and why “private by default” is about to matter far more than it sounds. Paid subscribers get the full teardown.

Subscribe now

——————————————————————————————————————

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, right now.

Get it Here

——————————————————————————————————————

Next week’s deep-dive:

——————————————————————————————————————

Five rounds this week added up to roughly $1.31 billion, and four of the five had AI somewhere in the pitch. But the real through-line isn’t “AI is hot.” It’s that the money is now paying for deployment, not demos: real customers, real revenue, real clinics. Here are the five that matter, ranked.

1. Neko Health, $700M Series C Daniel Ek’s preventive-health company raised $700 million led by Lightspeed and O.G. Venture Partners at a valuation of nearly $7 billion, up roughly 4x from early 2025. Neko builds its own body-scanning hardware, software, and clinics in-house, has run over 100,000 scans in the UK and Sweden, and is opening its first US clinic in New York. Spence’s take: A $7 billion price tag on a £299 cash-pay scan is a bet on human nature, not hardware. And human nature has a losing record when it comes to paying to hear bad news early.

2. Chai Discovery, $400M Series C Chai raised $400 million led by Index Ventures at a $3.8 billion valuation, nearly tripling its price in seven months and pushing total funding above $630 million. It builds AI models that design antibodies from scratch, and the tech is already in use at Lilly, Pfizer, and Novartis. Spence’s take: When Big Pharma is running your models in live programs, you’ve stopped selling promise and started selling proof. That, not the demo, is what tripled the valuation.

3. Emergent, $130M Series C The Indian “vibe-coding” startup hit unicorn status at a $1.5 billion valuation on a $130 million round led by Creaegis, a 5x jump in six months. Founded in mid-2025, it lets non-technical people build full apps from plain English and reports around $120 million in annualized revenue. Spence’s take: The revenue is real; the moat is the open question. In a field already crowded with Replit, Lovable, and Cursor, the winner won’t be whoever codes best. It’ll be whoever owns the non-technical builder first.

4. Oak, $60M seed Oak came out of stealth with a $60 million seed co-led by Accel, CRV, and Greylock to build an “identity operating system” that governs every human, machine, and AI agent inside a company. Founder Shai Morag has sold three prior security companies. Spence’s take: $60 million is an enormous seed, and it’s a signal. Every AI agent you deploy is a new hire with system access and no manager. Oak is building the badge reader for a workforce that isn’t human, and that problem is about to be everywhere.

5. Cyclops, $20M Series A The smallest round of the week might be the sturdiest. Cyclops raised $20 million (Nava Ventures led, with Circle and Coinbase Ventures in) to sell stablecoin payment plumbing to the payments industry itself. Shift4 and Mastercard are customers; the company reports over $2 billion in volume and more than 100 licenses. Spence’s take: Selling shovels to the whole industry beats selling gold to consumers, and 100-plus licenses is a wall, not a line item. It’s the same regulated-markets logic we build around at /mkt.

The #1 story, and the model to read it with

Neko is the round of the week, and the right lens is Incentives. Charlie Munger liked to say that if you show him the incentive, he’ll show you the outcome. The US healthcare system is paid to treat sickness, not prevent it. The money moves once you’re already sick, which is exactly why no incumbent has built Neko and, structurally, can’t. Neko routes around the broken incentive by going cash-pay, direct to the consumer. The bull case: it owns a market incumbents are financially discouraged from entering. The bear case is that same incentive in reverse. Insurers don’t pay for prevention, so Neko has to prove that enough people will spend their own money, at scale, to find problems they’d sometimes rather not know about. Owning the full stack is what lets it control the price and the experience. Whether that clears US regulation and insurance dynamics is the thing to watch.

Next week on Startup Spotlight

Every company above made noise this week. Next Wednesday’s paid deep-dive goes the other direction: the biggest AI infrastructure bet of the moment might be the one that never touches the cloud. I’m breaking down the on-prem AI shift, who’s quietly winning it, and why “private by default” is about to matter far more than it sounds. Paid subscribers get the full teardown.

Subscribe now

——————————————————————————————————————

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, right now.

Get it Here

——————————————————————————————————————

Next week’s deep-dive:

——————————————————————————————————————

On June 30, the Premier Lacrosse League closed a $100 million Series E, the largest capital raise in the history of professional lacrosse. Ares Management funds and billionaire Joe Tsai led it.

Plain English on what the PLL is. It’s a professional men’s lacrosse league, eight teams, the best players in the world. Founded in 2018 by brothers Paul and Mike Rabil off a roughly $3 million seed round. Paul was an All-American at Johns Hopkins and one of the most decorated players the sport has produced. Mike is a former Dartmouth football player and the operator of the two. Mike is CEO, Paul is president and chief creative officer.

Keep this in mind about the founders: before the PLL, professional lacrosse in America was Major League Lacrosse, a struggling, conventionally structured league. The Rabils didn’t join it. They built a competitor, then acquired it, merging with MLL in 2020. They don’t take the field as it is.

The round brings total funding to more than $218 million per PitchBook. The cap table now includes Ares, Joe Tsai Sports, ESPN, Arctos, The Chernin Group, The Raine Group, CAA, and Blue Pool Capital. Ares’ Jim Miller, co-head of its Sports, Media and Entertainment group, joins a board that already includes Tsai, Arctos’ David O’Connor, Raine’s Colin Neville, Chernin’s Mike Kerns, and CAA’s Mike Levine.

The celebrity layer got the headlines: ESPN made a subsequent minority equity investment, and the round drew Glen Powell, Rob Mac (formerly McElhenney), country singer Warren Zeiders, and actor Tony Cavalero. Powell joins as a creative advisor to help launch the first PLL and WLL teams in Texas.

A person familiar with the deal told Bloomberg it values the league at more than $500 million. The Rabils declined to confirm any valuation, saying only that it’s up since the 2022 Series D. Treat the $500 million as reported and unconfirmed.

Read more

On June 25, Arca came out of stealth with $64 million. The $48.5 million Series A was led by General Catalyst, with Index Ventures and Venrock backing it. Venrock had led the earlier $15.5 million seed.

Here’s the plain-English version. Arca is a wealth management firm. Human advisors, real clients, real portfolios. The twist is that the back office runs on AI agents built in-house, and clients never touch them. Founder and CEO Rron Rexha told InvestmentNews that every Arca client gets a human advisor and doesn’t interact with the AI directly, because the service is built on trust and the human is what makes trust work.

The company reports over $1 billion in client assets with 28 employees. Offices in New York, Maryland, and New Hampshire. A reported minimum of $500,000 in investable assets to become a client.

Rexha is a former product leader at Plaid and an immigrant from Kosovo. He didn’t build the board you’d expect from a fintech founder. He built one you’d expect from an operator trying to buy credibility in a regulated business: Bill McNabb, the former CEO and chairman of Vanguard. Jason Wenk, founder and CEO of the custodian Altruist. Peter Crawford, formerly of Charles Schwab. Morgan Housel, who wrote The Psychology of Money. Alex Tran and Nat Levy-Westhead from General Catalyst. Nick Beim from Venrock.

Now the part that most of the coverage skipped.

Per Arca’s Form ADV, the firm was seeded through Granite Bay Wealth Management, a Portsmouth, New Hampshire RIA founded in 2015. Then, in May 2026, Arca acquired Sandbox Financial Partners, a Bethesda, Maryland RIA with roughly $682 million in AUM.

Add those two books together and you’re most of the way to “over $1 billion.”

RIABiz reported this and quoted a critic who put it bluntly: the growth story is acquisition, not AI. Michael Kitces, who knows this industry as well as anyone alive, framed the open question as whether and where the productivity actually shows up. Wenk pushed back and said Arca’s organic growth is substantial too, pointing to roughly $1.27 billion in client assets as of December 31, 2025.

I’ve read a lot of funding announcements. This is the first one in a while where the filings tell a more interesting story than the press release.

Read more

Quick note before we start. SambaNova’s $1 billion round got its own post this week, so it’s not on this list. What’s left is five deals that share a theme. Almost all of this week’s money went to helping people own their stack instead of renting it. Here’s what actually mattered.

1. Oratomic raised $300M to skip straight to fault-tolerant quantum

Pasadena-based Oratomic announced a $300 million Series A, co-led by ARCH Venture Partners, Spark Capital, and Khosla Ventures, per the company’s announcement and reporting from The Quantum Insider and Yahoo Finance. Vinod Khosla posted on X that it was his firm’s largest initial investment since OpenAI. The company says its neutral-atom approach could reach useful quantum computing with roughly 10,000 qubits instead of the millions long assumed, though that number is a company-affiliated research claim, not settled science.

Spence’s take: When Khosla name-checks his OpenAI bet, study the structure of the wager, not the odds.

Here’s that structure, through the lens of a convex bet, a model from my book. The downside is capped at the check: $300 million, a knowable number. The upside, if fault-tolerant quantum actually arrives this decade, runs close to uncapped, from drug discovery to breaking the encryption that secures the modern internet. You don’t make that bet because you’re sure it works. You make it because the payoff is so lopsided that being wrong costs you one round while being right rewrites an industry. That asymmetry is also why generalist funds that kept quantum at arm’s length for a decade suddenly showed up at the table.

2. Norm AI hit unicorn status attacking the billable hour

Norm AI raised a $120 million Series C at a $1.2 billion valuation led by Khosla Ventures, per the company and TechCrunch. It embeds law and regulation into AI agents and runs an affiliated AI-native law firm that, according to its own materials, prices work on outcomes rather than billable hours and serves clients representing more than $30 trillion in combined assets under management (a company-reported figure).

Spence’s take: The scariest competitor isn’t the one with a better product. It’s the one whose pricing model your incumbents can’t copy without blowing up their own P&L.

3. Prime Intellect raised $130M so companies can be their own AI lab

Prime Intellect closed a $130 million Series A at a reported $1 billion valuation, led by Radical Ventures with NVIDIA Ventures, Intel Capital, and Dell Technologies Capital participating, per the company and TechCrunch. The pitch is to give enterprises the tooling to train and run their own agents instead of renting frontier APIs. TechCrunch reports the company has reached a $100 million annualized revenue run rate (company-reported).

Spence’s take: “Rent versus own” is the oldest question in business. AI just made it expensive enough to matter again.

4. Ollama raised $65M with 14 employees

Ollama announced a $65 million Series B led by Theory Ventures, bringing total funding to $88 million, per the company’s announcement. The kicker isn’t the money, it’s the leverage: the company reports 8.9 million monthly developers and a presence in 85% of the Fortune 500, run by a team of 14.

Spence’s take: Fourteen people, most of the Fortune 500. That ratio is the whole AI thesis in one line.

5. 8090 Solutions raised $135M to build software with agents on a leash

8090 Solutions, co-founded and led by Chamath Palihapitiya, raised a $135 million Series A led by Salesforce Ventures, per Crunchbase. The company builds enterprise software using coordinated AI agents kept under human-led oversight, a bet that the enterprise buyer wants autonomy and a human hand on the wheel at the same time.

Spence’s take: “AI does the work, humans own the outcome” is going to be the enterprise default. Anyone still pretending it’s one or the other is selling to the wrong decade.

The thread

Quantum hardware, legal work, model training, local inference, enterprise software. Five different sectors, one instinct: own the layer that matters instead of renting it. When capital moves this consistently in one direction, it’s usually pricing in a belief about who holds the leverage next.

Next week’s deep-dive goes straight at that shift. We’re digging into why the most regulated buyers on earth are pulling AI inside their own walls, and what the on-prem move means for anyone building in a compliance-heavy market. It’s the quiet story under half the rounds above.——————————————————————————————————————

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, right now.

Get it Here

——————————————————————————————————————

Next week’s deep-dive:

——————————————————————————————————————

1. Venture Capital Just Had Its Best Half-Year on Record. Two Companies Ate Half of It.

Global venture funding reached a record $510 billion in the first half of 2026, surpassing the $440 billion invested in all of 2025 and setting a new high for startup investment in any half-year period on record. OpenAI and Anthropic alone accounted for $217 billion 43% of all startup funding in H1. Strip those two out and the market looks dramatically different than the headline suggests.

Spence’s take: Record funding with record concentration isn’t a healthy market. It’s a market with one very loud story and a lot of everyone else fighting for the scraps.

2. Joulent Just Got $1.75B to Power Microsoft’s AI Before the Grid Can

National Grid Ventures agreed to invest $1.75 billion for a 35% stake in Joulent, forming a strategic partnership to develop contracted power and electrical infrastructure solutions for large load demand. The funding will support Joulent’s Project Kilby, a 2.67 GW co-located power facility in West Texas developed in partnership with Chevron, which will supply a Microsoft-operated data center campus under a 20-year power purchase agreement.

Spence’s take: Joulent launched in June and closed a $1.75 billion deal in July. The fastest path to a massive check is starting with a 20-year customer contract already in hand.

3. Together AI Raised $800M and Nobody’s Talking About What It Actually Signals

Together AI’s latest round confirms that AI compute is a must-own sector, with the neocloud model providing GPU power on demand for running and fine-tuning AI addressing a core bottleneck. The valuation jumped from $3.3B at Series B to $8.3B in this Series C led by Aramco Ventures, with NVIDIA, General Catalyst, Vista Equity Partners, and Emergence Capital all participating.

Spence’s take: When Aramco leads an AI infrastructure round, that’s not a tech bet. That’s an oil nation hedging its energy dominance with compute dominance. Different asset, same instinct.

4. Quantum Systems Raised $1.2B and Europe Finally Has a Defense-Tech Story Worth Watching

German drone maker Quantum Systems raised $1.2 billion in a Series D round, valuing the company at roughly $8 billion, backed by investors including Blackstone, Airbus, Advent, and Noteus. The round is one of Europe’s largest defense-tech financings ever and reflects a fundamental shift: European governments are no longer waiting on legacy contractors to move fast.

Spence’s take: Ukraine proved that cheap, fast, autonomous systems beat expensive slow ones on the battlefield. A $1.2 billion drone startup is just that thesis with a balance sheet.

5. Venice Raised $65M to Sell AI Privacy at a $1B Valuation and the Timing Is Perfect

Venice, developer of a platform enabling private, surveillance-free access to a wide array of AI models, secured $65 million in Series A funding led by Dragonfly, setting a $1 billion valuation for the 2-year-old Sheridan, Wyoming-based startup. The round lands exactly as enterprise concerns about AI data retention and model training on proprietary inputs are hitting a fever pitch.

Spence’s take: Privacy used to be a feature people said they wanted and didn’t pay for. Enterprise compliance is making it a line item they can’t avoid. That’s a different kind of demand.

Why Joulent Is the Story of the Week

I want to run Joulent through constraint mapping, a model from Mental Models: How to Think, Act, and Win. Every major technology wave eventually hits a physical constraint that software can’t solve. For AI, that constraint is power. Not chips. Not models. Actual electrons delivered at scale and speed. AI-related electricity demand is reshaping energy markets globally, with data centre demand rising 17% in 2025 against 3% growth in overall global electricity demand. Joulent didn’t try to compete in the crowded AI software market. It found the constraint that blocks everyone in that market and built the infrastructure to remove it. That’s the most durable position in any technology cycle: being the thing everyone else needs to exist.

Next week I’m going deep on one company from this list. My early pick is the story that looks like an energy deal but is actually a 20-year bet on who controls AI compute infrastructure. Subscribe so you don’t miss it.

——————————————————————————————————————

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, right now.

Get it Here

——————————————————————————————————————

Next week’s deep-dive:

——————————————————————————————————————

1. Menlo Ventures Just Cashed In a $500M Bet Nobody Else Wanted to Make

Menlo Ventures raised $3 billion in new capital, the largest fundraise in the firm’s 50-year history, after its early investment in Anthropic, now worth nearly $14 billion as Anthropic’s valuation climbed past $900 billion. Menlo led Anthropic’s Series D in 2024 by structuring roughly $500 million of that round as a special purpose vehicle, at a time when no major firm was writing three-quarter-billion-dollar checks. AI Funding TrackerTech Funding News

Spence’s take: Everyone calls these bets “contrarian” after they pay off. The actual contrarian move happened in 2024, when nobody knew if it would work.

2. XCures Doubled Its Valuation Cleaning Up Medical Records With AI

XCures landed a $46 million Series B led by Innovius Capital, bringing total funding to more than $76 million and valuing the company at $127 million post-money, more than double its prior round. The company grew from roughly $3 million to $10 million in annualized recurring revenue in 2025 and is on track to break $20 million in 2026, while achieving cash-flow breakeven last year. Calcali Tech

Spence’s take: A profitable company intentionally entering a burn phase to grow faster is one of the most underrated signals in venture. It means the founders chose growth, not the other way around.

3. Healthcare AI Just Crossed an $8.5 Billion Threshold This Year Alone

Investors have put an estimated $8.5 billion into seed- to growth-stage funding for AI-powered health tech companies in 2026 as of late June. Calcali Tech

Spence’s take: Healthcare moves slower than every other sector until it doesn’t. This number says we’re past the “if” stage and into the “how fast” stage.

4. Twenty’s $1B Cyber Warfare Bet Is Still the Story Nobody’s Fully Priced In

Twenty, the Arlington-based offensive cyber startup for the U.S. military, became America’s first venture-backed cyber warfare unicorn earlier this month after raising a $100 million Series B at a $1 billion valuation, less than two years after founding.

Spence’s take: A billion-dollar valuation with no disclosed revenue isn’t proof of a great company. It’s proof that investors believe the category just became fundable. Those are very different things.

5. Menlo’s Anthology Fund Shows the New VC Playbook: Be the Insider, Not Just the Investor

Menlo launched a $100 million fund called Anthology in partnership with Anthropic, which has since grown to roughly $250 million deployed, backed more than 60 companies, and produced exits including Graphite (acquired by Cursor) and Astrix Security (acquired by Cisco). Tech Funding News

Spence’s take: The smartest funds aren’t just writing checks anymore. They’re building proprietary pipelines into the company they bet the firm on. That’s a moat most LPs don’t fully appreciate yet.

Why Menlo Is the One to Watch

I want to run this through asymmetric bets, a model I cover in my book, Mental Models: How to Think, Act, and Win. Menlo’s 2024 Anthropic investment had a defined downside: the capital they put in. The upside, if Anthropic became one of the most valuable private companies in the world, was essentially unbounded. That’s the entire logic of venture capital distilled into one decision. Most investors talk about asymmetric bets. Menlo actually structured a $500 million SPV around one, at a moment when the broader market had gone quiet on big checks. The lesson isn’t “go invest in AI labs.” It’s that real conviction shows up as structuring your bet correctly when consensus is against you, not just picking the right company.

——————————————————————————————————————

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, right now.

Get it Here

——————————————————————————————————————

Next week’s deep-dive:

——————————————————————————————————————

1. Odyssey Just Raised $310M to Teach AI How the Physical World Actually Works

Odyssey, a world model AI startup founded by self-driving vehicle veterans Oliver Cameron and Jeff Hawke, raised a $310 million Series B at a $1.45 billion valuation, led by Natural Capital with Amazon, AMD Ventures, GV, and others participating. Instead of predicting the next word in a sentence, the company’s models learn to predict how the physical world behaves, including how objects move and what happens after something is touched or pushed. Odyssey’s CEO says the field is approaching the “GPT-3 moment” for world models, the point where the technology shifts from promising research into a breakthrough foundational layer.

Spence’s take: Everyone’s been racing to make AI talk better. This round is a bet that the next unlock is making AI understand physics better. Different race, same stakes.

2. SiFive’s $400M Says the Chip Wars Aren’t Just an Nvidia Story

Semiconductor startup SiFive raised a $400 million Series G round led by Atreides Management, challenging incumbent Arm Holdings with chip designs built on an open rather than proprietary standard.

Spence’s take: Open standards versus proprietary lock-in is one of the oldest fights in tech. SiFive just got $400 million more ammunition to fight it in chips.

3. Capital Is Concentrating Around Problems Nobody Can Commoditize

This week’s funding tape shows investors concentrating capital around a narrow set of hard-to-commoditize problems: sovereign cyber defense, enterprise workflow automation with measurable ROI, custom silicon design, and infrastructure sitting close to the AI compute bottleneck. Deals like Dream’s $260 million round, Gradial’s $65 million Series C, and Architect Labs’ $24 million seed all point to the same investor instinct: back teams attacking expensive, mission-critical systems where incumbents are slow to adapt.

Spence’s take: When investors stop chasing novelty and start chasing bottlenecks, that’s usually a sign the easy money phase of a cycle is ending.

4. Odyssey’s Amazon Deal Is the Real Story Buried in the Headline

Alongside the funding, AWS became Odyssey’s preferred cloud provider, with the startup committing to optimize its models for Amazon’s Trainium AI chips, positioning them as an alternative to Nvidia’s hardware.

Spence’s take: Funding rounds get headlines. Cloud and chip commitments get ignored. This one’s worth watching, because it’s Amazon picking a side in the AI infrastructure fight before most people noticed there was a fight.

5. The Smartest Money in the Room Showed Up as Individual Investors, Not Funds

Odyssey’s investor list includes existing backers like Jeff Dean, Google’s chief scientist, Elad Gil, Qasar Younis, co-founder and CEO of Applied Intuition, Garry Tan, president and CEO of Y Combinator, Guillermo Rauch, founder and CEO of Vercel, and Kyle Vogt, founder of Cruise.

Spence’s take: When the people building competing AI companies personally write checks into a rival, that’s not just a vote of confidence. That’s insiders hedging their own thesis.

Why Odyssey Is the One to Watch

I want to run this through second-order thinking, a model I cover in my book, Mental Models: How to Think, Act, and Win. The first-order story is “AI startup raises $310 million.” The second-order story is what happens if world models actually work: robotics, gaming, autonomous systems, and scientific simulation all get access to AI that understands cause and effect in the physical world, not just language patterns. That’s a bigger shift than a better chatbot. It’s AI gaining a working model of reality itself, and the company that gets there first doesn’t just win a product category. It potentially becomes the infrastructure layer everyone else builds on top of.

——————————————————————————————————————

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, you can grab my book, Mental Models: How to Think, Act, and Win, right now.

Get it Here

——————————————————————————————————————

Next week’s deep-dive:

——————————————————————————————————————

1. Bezos Bets $12B That AI Can Replace the Slowest Part of Engineering

Prometheus raised $12 billion in a Series B from JPMorgan Chase, Goldman Sachs, BlackRock, and Bezos himself, pushing total funding past $18 billion after launching just seven months ago. That puts the company’s valuation at roughly $41 billion, placing it among the top five most valuable AI startups on the planet. The 150-person company is building an “artificial general engineer” designed to compress the design-to-manufacturing cycle for physical products from years down to months.

Spence’s take: Seven months, $18 billion, zero disclosed revenue. This is either the smartest bet in physical AI or the most expensive science project in history. There’s no third option.

2. Robinhood’s Prediction Markets Playbook Just Got a $1.4B Validation in Germany

NEURA Robotics, a German physical AI robotics company, raised up to $1.4 billion in Series C funding led by Tether, alongside Qualcomm, Amazon, and NVIDIA, marking the largest round ever for a full-stack robotics company.

Spence’s take: When crypto-native capital and chip giants write the same check, that’s not hype. That’s two completely different risk appetites agreeing on the same thesis.

3. The Gaming Industry Just Bought a Piece of Its Own Plumbing

Interchecks, an instant payments platform serving sportsbooks and fintechs, raised $50 million in a Series C led by Bettor Capital, a growth fund focused entirely on real-money online gaming, with Commerce Ventures, Decades Holdings, and Thayer Street Partners also joining. The company has processed more than $50 billion in transactions over a decade, with triple-digit revenue growth for seven straight years and profitability since 2023.

Spence’s take: A profitable, decade-old company raising venture money on its own terms is rarer than the unicorn headlines. Pay attention when boring and profitable decide to raise anyway.

4. Voice AI Is Crowded, But Investors Still See Room for a Winner

Bland closed a $50 million Series C led by Dell Technologies Capital, with HubSpot Ventures and others joining. The company says it handles more than 3.5 million calls per week and has processed over 175 million AI phone calls total.

Spence’s take: Crowded markets don’t scare good investors. They just raise the bar for who gets funded next.

5. Wall Street’s Biggest Names Just Quietly Bet on Tokenized Assets

Receipts Depositary Corporation announced a $7 million oversubscribed round led by LiveOak Ventures, with Hivemind Capital, OTC Markets Group, and GTS also participating, to issue depositary receipts on digital and alternative assets.

Spence’s take: Small check, big signal. Traditional market infrastructure players are building the plumbing for tokenized assets before retail even notices. This is the kind of deal that looks irrelevant until it isn’t.

Why Prometheus Is the One to Watch

I want to run the Prometheus deal through second-order thinking, one of the models in my book. The first-order story is “AI startup raises a lot of money.” The second-order story is what happens if this actually works: entire categories of physical engineering jobs get compressed into a software workflow, and the bottleneck for inventing new jet engines, drug compounds, or hardware shifts from human design cycles to compute and data access. That’s not an incremental shift. That’s a complete rewrite of who controls the pace of physical innovation, and right now Bezos and a handful of banks are betting $41 billion that they’ll own that rewrite.

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, my book is out now.

Get it Here

Next week’s deep-dive:

Five rounds closed this week. Add them up and you get roughly $89 billion moving in a handful of days. That’s not noise. That’s a market telling you exactly what it wants, if you bother to read the tape instead of the headlines.

Here are the five that actually matter, and the one signal sitting underneath all of them.

1. SpaceX priced the largest IPO in history

SpaceX priced its IPO Thursday night at $135 a share and debuts on Nasdaq today under the ticker SPCX. The raise: about $75 billion across roughly 555.6 million shares, valuing the company near $1.77 trillion. For scale, the prior record holder, Saudi Aramco, raised $29.4 billion back in 2019. This deal is more than 2.5 times bigger. Musk keeps north of 82% voting control. Reported 2025 revenue was $18.7 billion, up 33% year over year, against a net loss near $4.9 billion.

Spence’s take: A rocket company is going public, but the asset doing the heavy lifting is Starlink’s recurring revenue. The hardware gets the headlines. The subscriptions pay the bills.

2. Bezos raised $12 billion for a startup that’s seven months old

Prometheus, the industrial AI company where Jeff Bezos is co-CEO, announced a $12 billion Series B at roughly a $41 billion valuation. The company launched in November 2025 with a $6.2 billion Series A, so its valuation has quadrupled in about seven months. Backers include JPMorgan, BlackRock, Goldman Sachs, DST Global, and Arch Venture Partners. Headcount: around 150 people. The pitch is an “artificial general engineer” that helps small teams design and manufacture physical things.

Spence’s take: $41 billion for 150 people is a bet on the founder and the category, not the cash flow. Sometimes that bet’s right. The check size tells you how badly nobody wants to miss it.

3. NEURA Robotics landed the biggest full-stack robotics round ever

Germany’s NEURA Robotics announced a Series C of up to $1.4 billion at a reported ~$7 billion valuation, led by stablecoin issuer Tether, with Nvidia, Amazon, Qualcomm, Bosch, Schaeffler, and the European Investment Bank alongside. The company raised $55 million in 2023, so this is a roughly 25x jump in capital in three years. Read the word “up to” carefully: the full amount is tied to performance milestones. It’s a ceiling, not a wire transfer.

Spence’s take: The investor list maps almost perfectly onto the supply chain a humanoid business needs to exist. The money’s strategic before it’s financial.

4. Digital Asset pulled $355 million into regulated-finance rails

New York’s Digital Asset closed about $355 million this week, led by a16z crypto with Citi Ventures, Dragoneer, Accel, and 70-plus other investors, pushing total funding to roughly $500 million. The product is blockchain infrastructure for regulated financial markets. Less flashy than a rocket. Arguably stickier.

Spence’s take: Building in regulated markets is slow, unglamorous, and brutally hard to copy once you’ve done the compliance work. It’s the same reason a platform like /mkt runs Reg A+ offerings through established trading infrastructure rather than reinventing the rails. The moat is the paperwork nobody else wants to do.

5. PhysicsX raised $300 million to simulate the physical world

London’s PhysicsX closed a $300 million Series C at about a $2.4 billion valuation, led by Singapore’s Temasek, with NVIDIA, Siemens, Applied Materials, and Atomico in the mix. The company builds AI that simulates physics for engineering and hardware design.

Spence’s take: Notice who’s investing. Siemens and Applied Materials don’t write checks for vibes. When your customers fund you, the round is also a contract.

The signal underneath all five

Look at the list again. Rockets. Industrial engineering AI. Humanoid robots. Regulated finance rails. Physics simulation. Not a single pure software app in the bunch. Capital this week ran straight at companies fusing software with hard, physical, or regulated systems. The “AI wrapper” era is over. Atoms and regulation are back.

Going deeper on #1: the Margin of Safety

The mental model worth applying to the SpaceX debut is Margin of Safety, the gap between what you pay and what a thing is worth. At a $1.77 trillion valuation against $18.7 billion in revenue, the price works out to roughly 95 times sales, with a real net loss underneath it. That doesn’t make it a bad company. It makes the price one that leaves very little room for error.

Pair that with a deliberately thin float and a fixed offer price, and you get a setup primed for sharp moves in either direction. The model isn’t a verdict on SpaceX. It’s a discipline: when price runs far ahead of demonstrated value, your cushion is thin, and thin cushions punish surprises. Treat the opening sessions as price discovery, not a referendum. The first earnings report will start settling the argument the offer price can’t.

That’s not a call to do anything. It’s a lens for reading what’s in front of you.

If this was useful, share it with someone who builds things. And if you want the full toolkit of 50 mental models, my book is coming soon.

Next week’s deep-dive:

1. Anthropic Is Raising $50 Billion at a $900 Billion Valuation — And the Round Is Almost Closed

Anthropic asked investors to submit allocations within 48 hours last week for what is expected to be its final private round before an IPO, targeting roughly $50 billion at a valuation approaching $900 billion. For context: Anthropic raised at a $380 billion valuation in February — meaning this round more than doubles that in under three months, driven by a revenue run rate that has climbed from $9 billion at year-end 2025 to closer to $40 billion today.

Spence's take:

Revenue going 4x in four months earns a valuation doubling. The number isn't the story. The trajectory is.

Mental model: Compound Growth.

This is what exponential actually looks like in real time — not the hockey stick on a deck, but $9B to $40B in annualized revenue inside one calendar year. Over 1,000 enterprise clients are now spending more than $1 million a year each on Anthropic's AI services. That's not a consumer product with viral distribution. That's enterprise compounding: each new customer validates the next, each deployment generates proprietary data, and the cost of switching to a competitor grows with every integrated workflow. The moat isn't the model. It's the installed base building on top of it. Once that flywheel reaches critical mass, the valuation math starts looking less insane than it does on a spreadsheet.

2. Parker Files Chapter 7. $200M in Funding. Gone.

Parker, a YC-backed corporate credit card startup for e-commerce businesses, filed for Chapter 7 bankruptcy on May 7, citing assets and liabilities each between $50 million and $100 million. The company had raised more than $200 million in total funding, including a $125 million lending facility, and reportedly hit $65 million in revenue before going dark.

Spence's take:

$65 million in revenue and a Chapter 7 filing in the same week is the clearest possible proof that revenue and unit economics are not the same scoreboard.

3. Big Tech Is Spending $725 Billion in Capex This Year — While Cutting Headcount

Meta, Amazon, Microsoft, and Alphabet have collectively signaled roughly $725 billion in capital expenditures for 2026, almost entirely earmarked for data centers, chips, and AI models, a more than 75% increase year-over-year. At the same time, Meta plans to cut 8,000 employees in May, Amazon has shed roughly 30,000 roles in recent months, and Microsoft has offered voluntary buyouts to about 125,000 employees.

Spence's take:

The pattern is simple: more capital into infrastructure, less capital into headcount. The companies that built AI tools to automate operational work aren't being hypocritical. They're being consistent.

4. Ramp Is in Talks for $750M at $40B — Six Months After Closing at $32B

Ramp is in advanced talks to raise $750 million at a pre-money valuation exceeding $40 billion, co-led by GIC and Iconiq Capital. The company crossed $1 billion in annualized revenue in 2025 and has been raising at step-up valuations multiple times per year. Meanwhile, Brex — its nearest competitor — was acquired by Capital One earlier this year for $5.15 billion, roughly 60% below its peak valuation.

Spence's take:

Same market, same product category, two completely different outcomes. Ramp built moats. Brex built scale. The difference compounds fast when the environment gets competitive.

5. Iridius Raises $8.6M to Embed Compliance Directly Into AI Workflows

Seattle-based Iridius raised $8.6 million in seed funding led by Chalfen Ventures with Accenture Ventures as both investor and strategic partner, building a platform that transforms regulatory standards into machine-readable executable logic and embeds them directly into enterprise AI workflows — so compliance happens continuously as systems run, not as a retrospective audit layer.

Spence's take:

Most compliance software manages paperwork after the fact. Iridius is building the layer that makes paperwork unnecessary. That's a fundamentally different bet — and Accenture writing a check alongside strategic partnership tells you the market is real.

1. The AlphaGo Guy Just Raised $1.1B — Before Writing a Single Line of Product Code

Read more