Trillion Dollar Baby

Anthropic sets IPO price range of $70 to $78 per share to reach an implied valuation of $1.404 trillion

Alex Shannon: I keep going back and forth on this — I think I actually land on the side that this is a good thing. Like, genuinely. A trillion-dollar-plus valuation for Anthropic, founders cashing out, the whole thing going public — I think that’s healthy for the industry.

Sam Hinton: Really? Because I read the same story and I came out the other end deeply uncomfortable. And it’s not the IPO itself — it’s the number. A trillion four. That’s not a valuation, Alex. That’s a belief system.

Alex Shannon: But belief systems are where all great companies start. Amazon was a ridiculous valuation for a bookstore. The question is whether the belief is grounded in something real.

Sam Hinton: And that’s exactly where I get stuck. Is it grounded in revenue? In customers? In some moat that I’m not seeing? Or is it grounded in vibes and venture capital momentum? Because those are very different foundations.

Alex Shannon: We are going to dig into every angle of this. But first — let’s get you properly oriented.

Alex Shannon: You’re listening to Build By AI — I’m Alex Shannon, and yes, that argument was real, and yes, we’re picking it back up in about sixty seconds.

Sam Hinton: And I’m Sam Hinton. Quick note before we dive in — starting today, Build By AI is moving to a weekly format. One bigger episode every Wednesday, covering the whole week’s AI news. Same great conversation, more depth, more stories. We think you’re going to love it.

Alex Shannon: It felt like the right move. The daily format was great but there’s something to be said for letting a week’s worth of stories breathe together — you start to see patterns you’d miss day to day.

Sam Hinton: And this week? There are patterns everywhere. We’ve got the Anthropic IPO, New York’s new AI safety law, OpenAI buying a travel startup for reasons no one will explain, Adobe going full aggressive on the creative tools front — it’s a lot. Let’s get into it.

Anthropic sets IPO price range of $70 to $78 per share to reach an implied valuation of $1.404 trillion

Alex Shannon: Alright, so the big story of the week — and honestly maybe the big story of the AI year — early reports suggest Anthropic is setting its IPO price range at $70 to $78 per share. At the top of that range, we’re talking about an implied valuation of $1.404 trillion on a fully diluted basis. That is a T. Trillion. With a T.

Sam Hinton: For context — that would put Anthropic in the conversation with some of the most valuable companies in human history. At that valuation, you’re in the neighborhood of companies that make physical things, sell physical things, have decades of revenue history. Anthropic is, what, a few years old?

Alex Shannon: Founded in 2021. And look, if confirmed, this is a company going public at a valuation that most established tech giants took decades to reach. So I’ll ask you directly — is that crazy, or is it justified?

Sam Hinton: Here’s my honest answer — it might be both. And that’s what makes it so interesting and so unsettling at the same time. The AI moment is real. The demand for AI infrastructure is real. Claude is genuinely competitive, arguably best-in-class in certain areas. So the underlying business has real substance.

Alex Shannon: But the multiple is extraordinary. Like, even if you believe everything good about Anthropic — the product, the team, the safety-first positioning — that’s still a staggering number to put on it. What do investors know, or think they know, that’s driving them here?

Sam Hinton: I think what they’re pricing in is the winner-take-most hypothesis. The idea that there will be maybe two or three foundational AI companies that everything else runs on top of, and Anthropic is positioning itself as one of them. If you believe that, a trillion-dollar valuation starts to make a weird kind of sense. It’s not crazy — it’s just very high conviction.

Alex Shannon: OK but here’s where I want to push back on your pushback from the cold open — because you said you were deeply uncomfortable. I get the skepticism on the number. But going public is actually a form of accountability, right? When you’re a private company, you can kind of hide the ball. When you’re public, you have to show your cards every quarter.

Sam Hinton: That’s a fair point, and I’ll give you that. There’s a transparency argument here that I actually buy. Public markets force a kind of discipline that private funding doesn’t. You can’t just raise another round at a higher number based on a vision deck. You have to perform.

Alex Shannon: Exactly. And maybe that’s healthy for AI as an industry. We’ve been in this phase where valuation has been almost fictional — just a number that one investor and one company agreed on in a room. Going public means the market gets to weigh in.

Sam Hinton: The market being a collection of people who also thought certain crypto tokens were worth billions, so I’m not sure they’re the ultimate arbiters of sanity. But yes — structurally, public markets are more accountable than private ones. I’ll concede that.

Alex Shannon: So where does that leave you? Still uncomfortable, or are you talking yourself into it?

Sam Hinton: Cautiously watching. The number is extraordinary. If that valuation is confirmed and the IPO prices at the top of range, Anthropic will have one of the highest-stakes stock debuts in tech history. And the pressure that creates on the business — to grow into that number, to keep innovating — that pressure is immense. I want to see if they can carry it.

Alex Shannon: Keep an eye on this because — and I mean this — the Anthropic IPO isn’t just a business story. It’s a signal about what the market thinks AI is worth right now. And that signal affects every company, every startup, every developer in this space.

All 7 Anthropic founders will each cash out about $250M in IPO

Alex Shannon: Now, you can’t talk about the Anthropic IPO without talking about what happens to the people behind it. And according to early reports from TechCrunch, all seven Anthropic co-founders are planning to sell shares in the IPO — each one receiving approximately $250 million at the midpoint price. And critically, if confirmed, they’re each selling the same number of shares. Equal payout across all seven founders.

Sam Hinton: The equal shares thing is actually the detail I find most interesting there. That’s unusual. Usually in a founding team there’s a clear hierarchy of equity — the CEO has significantly more, early contributors have more. The fact that they’re structured so that everyone gets the same payout in the IPO says something intentional about how this team is built.

Alex Shannon: It does feel deliberate. Like, that’s not an accident — that’s a cultural statement. And remember, this is a company that positions itself heavily around safety and mission. So the narrative they’re telling is: we’re a team, we’re aligned, no one person is bigger than the company.

Sam Hinton: Which I appreciate on a values level. But let’s also just acknowledge — $250 million each is an extraordinary amount of money. Seven people, $250 million each. That’s $1.75 billion flowing out to the founding team in this one event. That is life-changing, generationally-changing wealth.

Alex Shannon: And some people will look at that and say — good for them, they built something valuable. Other people will look at that and have questions about whether a company this laser-focused on the mission of safe AI should have founders walking out with that kind of payday. Is there tension there?

Sam Hinton: I think there’s a version of that tension that’s unfair and a version that’s legitimate. The unfair version is — why should anyone who built something successful get rich? That’s just how startups work. The legitimate version is specifically about Anthropic’s positioning. They’ve consistently framed themselves as the responsible, safety-first alternative. There’s a question of whether a $250 million personal payday per founder is consistent with that framing or whether it starts to look more like any other tech company in a tux.

Alex Shannon: That’s well put. The tux line is good. But doesn’t the fact that they’re maintaining control of the company soften that a bit? Early reports say the founders are retaining control even after the IPO. So this isn’t a cashing-out-and-leaving story — it’s a cashing-out-and-staying story.

Sam Hinton: It does soften it. If the founders were selling everything and stepping back, that would be a very different signal. The fact that they’re retaining control suggests they’re still committed to steering this thing. The $250 million is more like — taking some chips off the table because they’ve earned it, not because they’re done.

Alex Shannon: Which is honestly just smart financial planning. If 90% of your net worth is locked in one company’s stock, diversifying makes sense regardless of mission or values.

Sam Hinton: One hundred percent. And the broader point here for people watching the AI industry — this is what a maturing AI company looks like. The founders are human beings who built something, and they’re being rewarded for it. Keep an eye on whether the mission stays intact post-IPO, because that’s the real test. Right now, everything looks good. The pressure comes later.

OpenAI, Anthropic face new oversight as New York’s AI safety law takes effect

Alex Shannon: Alright, let’s stay in the Anthropic zone for one more story but shift the angle completely. While Anthropic is planning this massive IPO, they’re also — along with OpenAI — now subject to a new regulatory framework. According to early reports, New York’s AI safety law is now being enforced. And this is a real law with real teeth.

Sam Hinton: Yeah, this is a significant moment. We’ve talked a lot about AI regulation being discussed, being proposed, being debated — and here we are in New York where it’s actually in effect. Major AI developers have to register, disclose their safety practices, report critical incidents. And if they don’t — fines.

Alex Shannon: So let’s break down what that actually means in practice. If you’re OpenAI or Anthropic, you now have to tell New York what your safety practices look like. You have to have a process for reporting critical incidents. What counts as a critical incident here? Do we know?

Sam Hinton: The reports don’t get super granular on that definition, so I want to be careful not to speculate beyond what’s confirmed. But the structure — register, disclose, report, or face fines — that’s the architecture. And even without knowing every detail, that structure is meaningful because it creates accountability that didn’t exist before in a formal legal sense.

Alex Shannon: Here’s what I find fascinating about the timing. Anthropic is literally in the middle of an IPO process — which involves extraordinary levels of disclosure already — and simultaneously they’re now subject to this new New York law. Is there a tension there, or does one actually support the other?

Sam Hinton: Oh, that’s a great question. I actually think they might support each other more than they conflict. Because here’s the thing — a company going public wants to be able to say to investors: our safety practices are robust, they’re documented, they’re disclosed. If the New York law forces you to formalize all of that? That’s not a burden, that might actually be a selling point in the IPO prospectus.

Alex Shannon: Interesting reframe. So regulation as a credibility builder for investors. I hadn’t thought about it that way.

Sam Hinton: Exactly. And look — Anthropic in particular has staked its identity on safety. If you’ve been saying ‘we’re the safe AI company’ for years, being forced to prove that publicly in a regulatory filing is not a problem for you. It’s actually an opportunity. OpenAI’s relationship with that narrative is more complicated, so the calculus might be different for them.

Alex Shannon: Let’s talk about the bigger picture though. New York has always been kind of the financial and legal bellwether in America. When New York passes a law like this, other states notice. Other cities notice. Is this the beginning of a wave?

Sam Hinton: I think it accelerates one, yes. We’ve already seen various state-level AI regulation efforts. New York enforcing something — not just passing it but actually enforcing it with fines — sends a signal to the industry that the era of building first and asking for forgiveness later is closing. That window is not shut yet, but it is narrowing.

Alex Shannon: And for companies outside the AI frontier labs — smaller developers, startups, teams building on top of these models — does this affect them?

Sam Hinton: If the law targets ‘major developers’ specifically, smaller players may have more runway. But here’s the thing — once the big companies are required to disclose their safety practices, that creates a kind of benchmark. And investors, customers, enterprise clients — they’re going to start asking smaller companies: hey, what are your safety practices? The standard filters down even without a direct legal requirement.

Alex Shannon: The practical takeaway here — if you’re building in the AI space, start documenting your safety practices now. Not because New York law necessarily applies to you today, but because the expectation is coming. And you do not want to be scrambling to build that infrastructure when it’s legally required.

Anthropic launches Claude Meetings to turn workplace discussions into plans and tasks

Alex Shannon: OK let’s talk about something Anthropic is actually shipping right now, because amid all the IPO news, they also dropped a new product. It’s called Claude Meetings. According to TechCrunch, it’s a feature that uses Claude to turn workplace discussions into actionable plans and tasks. We’re talking drafting documents during meetings, mapping out timelines, answering questions in meeting chat, and accessing connected work apps to pull information.

Sam Hinton: I love this product direction and it’s also the most crowded space in AI right now. Every major AI company wants to be in your meeting. Microsoft has Copilot doing this in Teams, Google has Gemini doing this in Meet — and now Anthropic is coming in with Claude Meetings. This is the AI assistant battle being fought in conference rooms.

Alex Shannon: So the feature set sounds genuinely compelling — drafting plans during the discussion, mapping timelines on whiteboards, pulling info from connected apps. Walk me through what that actually looks like in practice. Like, if I’m in a product planning meeting, what does Claude Meetings actually do for me?

Sam Hinton: Think of it this way — you’re having a conversation about a project, and instead of someone frantically taking notes and then spending two hours turning those notes into a Jira ticket or a project brief, Claude is doing that in real time. The meeting itself becomes the input for the deliverable. That’s actually a meaningful shift in how meetings work.

Alex Shannon: Because right now, the meeting and the work product are separate things. You have the meeting, then someone has to translate it into something actionable. Claude Meetings is trying to collapse those two things into one.

Sam Hinton: Right. And the connected apps piece is crucial — if Claude can pull up the actual data from your work tools during the meeting, that closes another loop. How many times have you been in a meeting where someone says ‘I think we have about fifty users on that feature’ and nobody actually knows? Claude could just pull the number. No more ‘I’ll follow up on that.’

Alex Shannon: No more ‘I’ll follow up on that’ — that’s the dream, honestly. But I want to play devil’s advocate here because I’ve seen a lot of AI meeting tools promise exactly this and the reality is always a bit messier. The accuracy of real-time transcription, understanding context in a meeting, knowing when someone is joking versus when they’re making a real decision — those are hard problems.

Sam Hinton: Completely fair. And the quality of the output is everything here. If Claude generates a project plan that your team has to spend an hour correcting, you haven’t saved time — you’ve just moved the work. The value proposition only holds if the output is actually good and trustworthy.

Alex Shannon: So what’s the differentiator for Claude specifically? Why choose Claude Meetings over what Microsoft or Google is offering?

Sam Hinton: If you’re already a Claude shop — if Claude is the AI your team trusts and uses — this is a natural extension. And Anthropic has consistently gotten strong marks for the quality of Claude’s reasoning and writing. If those strengths translate to meeting outputs, that’s a real advantage. The question is whether companies want their AI ecosystem unified under one provider or whether they’re mixing and matching.

Alex Shannon: The practical implication for businesses watching this — the meeting room is becoming a major AI battleground. And the companies that figure out the right integration into how teams actually work will have a significant advantage. Keep an eye on what Claude Meetings actually looks like in real use, because the feature list sounds great but the execution is where it lives or dies.

Adobe brings AI agents and new creative tools to Photoshop, Premiere, and Firefly

Alex Shannon: Rapid fire time. First up — Adobe. Early reports suggest Adobe is bringing AI agents to Photoshop, Premiere, and Firefly, plus a new prototype builder tool that’s going directly after Figma. Sam, what do you make of it?

Sam Hinton: Adobe going after Figma is fascinating because — and everyone remembers this — Adobe tried to acquire Figma and that deal got blocked by regulators. So now they’re building a competitor instead. That’s a very specific kind of corporate frustration being channeled into product strategy.

Alex Shannon: They didn’t get to buy it so now they’re going to beat it. Classic.

Sam Hinton: And the AI agents across Photoshop and Premiere — this is Adobe trying to stay essential to creative professionals in a world where AI can generate images and video. They have to make the AI work for their existing users, not replace them. That’s a tricky balance and it’s the most important challenge Adobe faces right now. Watch how creative professionals actually adopt this — that’ll tell you everything.

OpenAI acquires Odessia, travel startup created by Sonder founder

Alex Shannon: Next — OpenAI acquired a travel startup. It’s called Odessia, founded by Francis Davidson, the same person who founded Sonder, the hospitality company. Early reports say OpenAI hasn’t announced what they’re doing with it. Sam — travel startup. What is OpenAI doing?

Sam Hinton: OK, so I’ve been thinking about this one and here’s my theory — travel is one of the most complex, most frustrating, most information-dense consumer experiences that exists. Booking a trip involves dozens of decisions, dozens of data sources, real-time information, personal preferences. If you’re trying to build an AI agent that handles real-world tasks for people, travel is a phenomenal training ground.

Alex Shannon: So it’s less about being a travel company and more about using travel as the domain to build agent capabilities.

Sam Hinton: That’s my read. And Francis Davidson built Sonder — he understands the operational complexity of hospitality at scale. That expertise plus OpenAI’s AI capabilities could produce something really interesting. Or OpenAI could do nothing with it and it fades quietly. We genuinely don’t know yet, and that uncertainty is part of what makes it worth watching.

Google broadens its AI hardware portfolio with two new specialized chips

Alex Shannon: Google hardware news — early reports indicate Google announced two new specialized tensor processing units, their AI chips. These ones support confidential model training and offer more memory capacity. Sam, chips. Why should a regular listener care about chips?

Sam Hinton: The confidential model training piece is the thing that jumped out at me. Enterprises have been cautious about training AI on proprietary data because of security concerns — what if the data is exposed? Chips that can support confidential training are essentially trying to remove that obstacle. That’s Google saying to enterprise customers — you can train on your sensitive data, we’ve got the hardware to protect it.

Alex Shannon: So it’s not just a spec upgrade, it’s addressing a specific business concern.

Sam Hinton: Exactly. And the more memory capacity matters because bigger, more capable models need more memory to run efficiently. Google building its own chips means they’re not dependent on Nvidia for their own AI infrastructure — that’s a strategic independence play that has enormous long-term implications. Every chip Google ships that’s good enough is a chip they don’t have to buy from someone else.

Oura to acquire AI-native primary care startup Counsel Health for $625M

Alex Shannon: Last rapid fire — Oura, the smart ring company, is acquiring Counsel Health, described as an AI-native primary care startup, for $625 million. And the kicker — Oura postponed its own IPO indefinitely, citing market uncertainty. Sam, Oura spending $625 million when it just punted on its IPO — what’s the story?

Sam Hinton: This is a fascinating strategic pivot. Oura postpones the IPO, which means they can’t raise public capital right now, but they’re making a massive acquisition anyway. That tells me they have enough private runway to do this and they see it as more valuable than waiting for the IPO window to open.

Alex Shannon: And the bet they’re making is that health data plus AI primary care is a combination that’s worth building right now, even in uncertain market conditions.

Sam Hinton: Think about what Oura has — continuous biometric data from people wearing their ring every day. Heart rate, sleep, temperature, activity. Now add an AI primary care layer on top of that. You’re not just tracking your health, you’re potentially getting care recommendations based on your actual real-time physiology. That’s a genuinely compelling vision for the future of healthcare. The question is whether regulators and consumers are ready for it — because that’s a lot of trust to extend to a smart ring company.

BIGGER PICTURE

Alex Shannon: Alright, let’s zoom out. We’ve covered a lot of ground today and I want to connect some dots because there are a couple of threads running through everything we talked about that I think are worth naming explicitly.

Sam Hinton: Yeah, go ahead — I think I see the same threads.

Alex Shannon: So the first one is obvious — Anthropic is everywhere today. IPO, founder payouts, New York regulation, Claude Meetings. But strip away the company name and what you’re actually seeing is the AI industry growing up. Going public, facing regulation, shipping product, compensating founders. That’s maturation. That’s not a startup anymore.

Sam Hinton: One hundred percent. And that maturation is happening simultaneously across the entire ecosystem. Adobe is integrating AI agents into tools that professional creatives have used for decades. Oura is extending AI into primary healthcare. OpenAI is buying a travel company. Google is building its own AI chips. These aren’t experiments anymore — these are strategic bets made by large, serious organizations with real money and real accountability.

Alex Shannon: And then there’s the regulation thread. New York’s AI safety law taking effect is not a minor story even though it might feel like a background detail compared to a trillion-dollar IPO. It’s actually the sign that the policy world has caught up enough to start drawing lines. And once one jurisdiction draws lines, others follow.

Sam Hinton: Here’s the pattern I’d want people to hold onto as they watch the next few months: we are moving from a world where AI companies were primarily accountable to their investors, to a world where they’re accountable to regulators, to customers, to public markets — all simultaneously. That’s a fundamentally different operating environment. The companies that thrive in it are the ones who built their governance and safety infrastructure early. The ones who didn’t are going to find this transition very uncomfortable.

Alex Shannon: And the question I keep coming back to — the one I want to leave you with — is this: at a trillion-dollar-plus valuation, with regulation arriving, with AI moving into meetings and healthcare and travel and creative work all at once, are we in a moment of genuine transformation? Or are we in an extraordinarily expensive and sophisticated bubble that’s going to teach us something hard about the difference between potential and reality?

Sam Hinton: My honest answer is — probably both. And the fascinating thing about being alive in this moment is that we don’t get to know which one it is until we’re already through it. What I’d say is: watch the revenue. Watch the customer retention. Watch whether the people using these tools are actually getting meaningfully better outcomes. That’s how you separate the transformation from the hype. And right now, there’s enough of both that it demands your attention.

OUTRO

Alex Shannon: That’s going to do it for this week’s Build By AI. This was our first weekly Wednesday episode and honestly — I think this format is going to be great. Having a full week of stories together gives the conversations so much more room to breathe.

Sam Hinton: Agreed. And there’s a lot coming — the Anthropic IPO is going to keep developing, the New York regulation story is going to unfold, we’ll see what OpenAI actually does with Odessia. We’ll be covering all of it next Wednesday.

Alex Shannon: If you haven’t subscribed yet, do it now wherever you get your podcasts — it’s the easiest way to make sure you don’t miss next Wednesday’s episode. And if this one was useful to you, share it with someone who’s trying to make sense of where AI is headed. We’ll see you next week.

Sam Hinton: See you Wednesday.