The U Lab Brief 21 | FAANG to MANGOS

The U Lab Brief 26 | Premium That Is a Discount

The most popular online payment processor on the planet just got a takeover bid at a discount to what payments businesses normally fetch. 439 million users. Roughly 44 percent of the global online payment market. And revenue still growing. And the market cheered. Here is the consensus. Stripe, alongside Advent International, has offered around $53 billion, according to Reuters. $60.50 a share, a 28 percent premium. PayPal closed up more than 17 percent. The press read it as a win, and the opening of a bidding war. Now the basics, because they carry the whole story. Every deal has two numbers, and they measure two different things. The premium tells you how far above today's price the buyer is paying. It says nothing about whether today's price was already low. The multiple ignores the stock price entirely. It asks a different question. What are you paying for the cash the business actually earns. This is why a bid can look generous and cheap at the same time. A large premium on a stock that already fell hard is still a low price for the business underneath it. The premium is measured against the fallen share price. The multiple is measured against the earnings. Two different baselines, so both readings hold at once. That is exactly the gap here. 28 percent over Tuesday looks like a gift to PayPal's shareholders. But 7 times EBITDA, against 8 to 12 times for payments peers, says the buyer is getting the cash flow at a discount. Both are true. The premium reflects a de-rated stock. The multiple reflects what the business earns. That is the whole point. So why does a cash machine trade below its peers? Because public markets price the narrative, not the installed base. PayPal's story soured. A post-pandemic reset, share lost to Stripe and Apple Pay, a tech stack patched together from years of acquisitions. Once the story breaks, the multiple compresses below what the cash flow is worth. Here is the model to remember. When a durable business loses its story, its public price falls faster than its cash generation does. That opens a gap between what the market pays and what the business earns. And that gap is an invitation. Note who accepts it. Stripe is the strategic buyer, Advent the financial sponsor. A strategic usually pays up for synergies. Yet even here the offer lands below peer multiples, which tells you how far this stock has fallen. That is the whole logic of buying a de-rated asset. You acquire durable cash flow the public market has stopped rewarding, and move it somewhere quarterly sentiment no longer sets the price. #PayPal #acquisition #MergersandAcquisitions #Fintech #PrivateEquity #shareprice #Stripe #AdventInternational #CashFlow #EPS

The U Lab Brief 25 | Paying More For Less

OpenAI wants to go public at $1 trillion, and this month it signaled it would rather wait until 2027 than list for a dollar less. So the question worth sitting with is simple: what would have to be true for that number to hold? Start with the scorecard. PitchBook rates the leading AI labs across five dimensions: revenue quality, capital efficiency, governance, moat durability, and how much of their own compute they control. OpenAI comes in at 4.53 out of 10. Anthropic at 8.2. On today's fundamentals, Anthropic grades higher. Now price that quality. Investors are paying about $188 billion for every point of OpenAI's quality, against $118 billion for Anthropic's. That's a 60% premium, for the company scoring lower. Which raises the real question: what are those investors seeing that the scorecard doesn't? The clearest place to watch that tension is the multiple. On its own valuation, the market grants OpenAI roughly 34 times revenue. At that multiple, OpenAI needs only about $29 billion in annual revenue to justify a trillion, well within reach. But apply Anthropic's multiple, around 20.5 times, and the same trillion suddenly demands $49 billion in revenue. Nearly double. So the whole trillion-dollar case comes down to one thing: whether public investors keep extending OpenAI the premium it enjoys in private markets. Around $340 billion of its $852 billion valuation rests on that richer multiple alone. Price it the way the market prices Anthropic, and that's the piece OpenAI has to earn. And waiting isn't free. On its own projections, OpenAI runs roughly $115 billion in cumulative losses before the business turns self-sustaining around 2030. But here's the other side, and it's a real one. That premium may not be irrational at all. 900 million people use OpenAI's products every week. The brand is the category. Investors aren't pricing the business it is, they're pricing the business it becomes. If OpenAI grows revenue into that 34-times multiple, a trillion doesn't look aggressive. It looks early. So don't watch the valuation, watch whether the fundamentals earn it. The price is already on the table. The only question left is whether OpenAI grows into it. Is that premium foresight or a bet? The S-1 will tell us. Source: PitchBook #VentureCapital #OpenAI #Anthropic #Technology #ArtificialIntelligence #AI #IPO #StartupValuation

The U Lab Brief 24 | Why Capital Concentrates

We can think of this as the liquidity-legitimacy loop. When liquidity becomes constrained, capital does not spread out. It moves toward the managers LPs already trust. The latest PitchBook NVCA Venture Monitor shows why. US venture exits reached $2.19 trillion in the first half of 2026, more than all venture exits of the previous decade combined. Yet distributions to LPs remain constrained. That distinction matters. Headline exit value is not the same as cash flowing back to limited partners. When distributions remain weak, LPs cannot keep making new commitments at the same pace without increasing their exposure to venture capital. So they consolidate. They commit to fewer funds, favour longer track records, and allocate more capital to managers who have already demonstrated returns. Experienced firms captured a record 89% of all VC fund commitments in the first half of the year. By fund count, they represented a record 62.2% of all funds closed. This is where the advantage begins to compound. Established firms already have longer track records and prior returns. That gives LPs more confidence in them. More confidence leads to more capital. And more capital gives those firms more chances to invest in the next generation of successful companies. Their past performance helps them raise more today, which can strengthen their position again tomorrow. Some emerging managers can still break through, particularly those with strong operating experience or established track records from previous firms. But the broader question remains: What happens to innovation when capital increasingly concentrates around the same venture firms? Many emerging managers invest at pre-seed and seed. They often provide the first institutional check, the capital that gives a young company the opportunity to move towards growth and scale. When liquidity tightens, prior performance becomes evidence of credibility. And when LPs rely more heavily on that evidence, established firms gain a larger share of new capital. So this is not only a fundraising problem for smaller funds. It may narrow the pipeline through which new innovative ideas receive their first institutional check. #VentureCapital #VC #Startups #LPs #LimitedPartners #EmergingManagers #Innovation #Fundraising #TheULab

The U Lab Brief 23 | The AI Deployment Layer

Enterprise AI may not be won by the company with the strongest model. It may be won by the company that can make AI work inside real organizations. According to TechCrunch, Microsoft’s new Frontier Company, backed by a $2.5 billion commitment and 6,000 industry and engineering experts, signals an important shift. The AI race is moving from capability to deployment. The question is no longer just: who can build intelligence? It is: who can translate intelligence into operating results? That is where Microsoft has a structural advantage: enterprise relationships, Fortune 500 presence, and the ability to bring AI into large, complex institutions. In enterprise AI, the next moat may not be the model alone. It may be the deployment layer. #AI #EnterpriseAI #Microsoft #VentureCapital #Technology #Innovation #TheULab

The U Lab Brief 22 | Engram: The Learned Memory Layer For AI

A startup called Engram has emerged from stealth with $98 million in funding from some of Silicon Valley's leading venture capital firms. Founded by researchers from Stanford, Berkeley, and Cornell, the company is already partnering with Microsoft, Notion, and Harvey. But the funding isn't the real story. The real story is that Engram is building what it calls a learned memory layer for AI. Today's AI is incredibly intelligent, but inside an enterprise it often behaves like a brilliant stranger. Every time it answers a question, it largely reconstructs an organization's context. It rereads documents, relearns processes, and rediscovers institutional knowledge again and again. As enterprises deploy AI agents across more functions, those repeated computations consume vast numbers of tokens, increase inference costs, and limit the efficiency of AI at scale. Engram takes a different approach. Instead of repeatedly retrieving information, its models study an organization's knowledge in advance and compress it into a compact, reusable memory. The longer the AI is used, the more it learns about the organization. According to the company, this allows its models to match or outperform frontier models while using up to 100 times fewer tokens, enabling faster responses, lower inference costs, stronger personalization, and more efficient long-running AI agents. One distinction is worth understanding. Conversation memory helps AI remember your interactions. Organizational memory helps AI understand your organization. The next competitive layer in enterprise AI may not be intelligence itself. It may be memory. Because intelligence answers questions. Memory compounds organizational knowledge. I'm Hurratul and this is The U Lab Daily Brief on venture capital, technology, and the future of innovation. Source: PR Newswire, StrictlyVC