Watching OpenAI promote its ads business puts me in something of a nostalgic mood. When I saw OpenAI’s thirsty announcement this week proclaiming a billion-dollar revenue run rate in its advertising business, I had myself something of a chuckle. We’ve seen this movie before – I wrote an early script for it in The Search, which among other things chronicled Google’s rise to a spectacular IPO in 2004. But is it really the same movie? The answer is no – but if you squint, and take a few reasonable liberties with history, it’s clear Altman & Co. are hoping for a sequel.

The Race to A Billion Trillion

For nearly my entire career, the tech industry was obsessed with billions. A tech unicorn is still defined as a startup valued at a billion dollars or more. Within that number lay bragging rights – and promises of massive paydays for LPs, entrepreneurs, and bankers.* But the goalposts have shifted over the past decade. A billion-dollar valuation is yesterday’s news. Now the goal is to hit a billion in revenue. If a startup is on track to reach that figure and can argue convincingly that it can grow indefinitely into the future, it has the potential to become tech’s newest and most sought after creature: a trillion-dollar “megacap” company.

Altman is hoping Wall St. will see in OpenAI what it saw in Google in 2004 and Facebook in 2014: An advertising-driven megacap just waiting to be born.

Both Google and Facebook made business history in their stock market debuts. Google’s S-1 announced search as the fastest growing business to ever hit the public markets.** Ten years later, Facebook rode social to the largest tech IPO to date (it raised $16 billion, which sure seemed like a lot back then…).

Now, thanks largely to the AI boom, three companies are rewriting the record books, all within the space of one year. SpaceX has already gone public and holds megacap status. Anthropic hopes to lap it with a multi-trillion dollar IPO later this Fall. But OpenAI, once considered the safest bet of the three, is struggling.

Lapped by Anthropic and facing unprecedented challenges – leadership turnover, AI safety crises, public backlash against data centers, a cratering bond market – OpenAI needs to paint as rosy a picture as possible to reach megacap status. Seen in that light, it’s no wonder the company decided to write a triumphal blog post about the growth of its ads business.

But is OpenAI’s milestone really that impressive? Yes – but with a few caveats. As the chart above illustrates, no matter how you cut it, OpenAI’s first year of advertising revenues are unequaled. The company hasn’t provided concrete figures, but hitting a billion-dollar run rate just six months after launch is an undeniably big deal. Based on that, I’ve estimated that OpenAI’s advertising platform will collect half a billion in revenue in its first full year. That’s more than three times Google’s (inflation-adjusted) $169 million in the first year after launching AdWords. And it’s more than six times the (again, inflation-adjusted) $80 million Facebook collected after launching its NewsFeed advertising platform. Remarkable.

But….wait, didn’t OpenAI promise $2.5 billion in ad revenues in its first year? Yup. It also promised $100 billion in year five. Let’s recalibrate and model OpenAI’s revenue based on Google and Facebook’s first five years of platform-based advertising revenue:

Still impressive, but even if OpenAI matches the first five years of Google’s blistering growth, it will fall far short of its original projections – $67 billion short, to be exact.

Now, one could argue that OpenAI may well grow far faster than Google did in its first five years. That would be a truly remarkable feat – but it would require that OpenAI steal significant share from both Google and Facebook. When Google launched AdWords and Facebook launched NewsFeed, they were creating entirely new markets. In 2001, less than five percent of corporate marketing budgets were allocated to digital. When NewsFeed launched in 2006, that figure had risen to a paltry six percent. Over the next decade, both companies benefitted from a once in a lifetime shift from analog to digital channels. Today, digital advertising comprises more than 70 percent of total spend. The transition is over – the winners are now incumbents.

OpenAI is launching an advertising platform into the headwinds of a duopoly that already owns more than 50 percent of digital ad spend. Add in Amazon and Apple, and that figure rises to 63 percent. Those four horsemen – Google, Meta, Apple, and Amazon – have far more data, capital, and experience that their upstart competitor. They’ll fight OpenAI at every turn. And that fight is existential. So while launching out of the gates with a billion-dollar run rate is certainly impressive, I think it’s fair to say that it’ll be …. challenging for OpenAI to grow as quickly as it – or Wall Street – would like it to.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

*And this is why, multiple times during my career as an entrepreneur, banks and VCs wooed my businesses with presentations and financing offers that suspiciously converged on a billion dollars. 

**There were faster growing companies by pure topline, but nearly all of them were financial engineering projects or companies with almost no revenue three years prior to IPOs.

A week ago I knocked my phone – a Google Pixel 8 Pro – off the table and watched helplessly as it bounced awkwardly onto the stone floor of our favorite restaurant. After retrieving it I noticed a splotch of dead pixels in the top right corner. This was a new phenomenon – I was expecting a cracked screen. But the display sported what looked like a deep bruise. Over the next few days, the black spot spread slowly toward the phone’s front facing camera. By midweek, the Pixel’s light sensor was enthusiastically involved, and my phone was rendered unusable.

I live on an island, so a quick repair or replacement is not an option. That’s why I keep my old phones in stock – it can take up to three weeks for a FedEx or UPS shipment to find its way to my home. I ordered up a replacement online (I settled on the Pixel 10, as the new 11 is on backorder for more than a month), fired up an old Pixel 6, and popped in my SIM.

It had been more than three years since I’d turned that phone on. A lot changes in that amount of time. I felt like a time traveler landing in the Spring of 2023. There were most of my familiar apps, but many were stuck in amber. Suddenly I was still halfway through The Covenant of Water on Audible. Long forgotten texts reminded me of relationships I’ve neglected to tend. My old Recount account was still integrated into my email app (we had just sold the company, and I was still consulting for the new owners). And apparently, the 2023 version of me thought Threads merited a place on my home screen. (That didn’t last long).

Of course, the most noticeable change was the absence of Claude, which since last year has sat prominently in the center of my home screen. I’m not entirely AI-pilled – I don’t use Claude as a frontend to a dedicated Mac Mini running endless loops of vibe-coded life hacks. But returning to a time before AI became the center of my technology life was….kind of nice.

Comparing the two vintages, separated by just a few years, has helped me understand what’s changed in my life – the habits I’ve picked up, what’s become more important, and what I’ve left behind. The Fidelity and Coinbase apps were buried in 2023, today I check them regularly. Why? I’ve decided to get serious about my far flung (and rather poorly managed) investments, and the best indication of that commitment is screen real estate. Also new: A folder dedicated to Google apps: Docs, Sheets, Slides and Calendar – we run DOC on Google, and DOC didn’t exist three years ago. Also new: Google Wallet, which I use to pay for pretty much everything. Three years ago, most of the stores on this rather quaint island were still using plastic. Now, they all take Apple or Google Pay.

I’m writing much more now, so WordPress has also moved into prime position. Lutron (a home automation app) and Oura have reappeared after many years offline – I’ve gotten more serious about my health, and I finally can control the damn lights in my house after many years of not having the time to figure it out. Constant throughout the past three years: Peloton, Maps, Slack, Strava, NYT Games, Spotify, and the Video and News folders. Relegated to second or third screens: Sonos, Firefox (used mostly for testing), and WhatsApp and Signal, which I used mostly for The Recount.

It’s been five or so days since my main phone went down, and perhaps the biggest change has been my commitment to the device itself. Because my old phone doesn’t have my new DOC email or Slack integrated, I’ve stopped bringing it with me when I go to dinner, a movie, or a weekend tennis match. That feels a bit like having a phantom limb – I’m so used to checking my phone whenever I have an idle moment. But I’m also enjoying the freedom of a tether-less life. Yesterday I took a two-hour walk and brought the old Pixel 6, sans SIM card. No calls, no texts, no email, no connectivity whatsoever. Just me, Audible, and a good book in my ears. I just might get used to this.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

August is often dismissed as a lost month in the business world – nothing gets done because everyone is on vacation. But where I live, August is the apex of the year. A heady mix of artists, financiers, philanthropists, business leaders and politicians converge on our little island for the month. The ensuing deluge of lecture series, film screenings, impromptu house concerts, and remarkable dinner conversations can be overwhelming. Four weeks into the month, those of us who live here are a bit strung out, ready for the tide of intellectual and cultural offerings to recede. Fall becomes a time to repair.

As might be expected, one topic dominated the conversations I’ve had at countless dinners, soirees, and fundraisers this summer. Yes, it’s “AI,” but that was true last summer, and even the summer of 2024. This year, however, the discussions turned from wonder and anticipation to something…darker. A central question came up over and over again: Is this technology – and the leadership behind it – fundamentally out of our control?

It’s not hard to understand why that question has become so dominant: The leaders of nearly all the frontier AI labs already have their answer. Here’s Elon Musk in a staff all hands this week, as reported by The Information yesterday:

“On the call, Musk said it is inevitable that AI models will become so advanced they’ll be impossible for humans to control. For that reason, he told employees … they needed to help SpaceX’s AI division build the technology before other companies did, the people said.”

Let that sink in. If you believe that AI will ultimately be out of control, then the best thing to do is to run the company that builds it first. That’s quite a performative, self-serving contradiction!

But self-serving statements are the coin of the realm at the highest levels of power in the Valley these days. And that’s what has my dinner companions in a dark mood. Here’s OpenAI CEO Sam Altman, who is neck-and-neck with Mark Zuckerberg in the race to say whatever helps him win: “The two biggest risks in AI today …. are loss of control, where AI becomes too powerful in a way that we can’t control, and the other is power gets too centralized.” This from the CEO of the second-largest pure play AI company in the world (OpenAI was #1 until they lost their lead to Anthropic a few months back. I guess by “centralized” he means… not him?)

But it’s not just the Musks and Altmans of the world that are driving my summer colleagues into a deep-seated pessimism. It’s also the inability of our current administration to step into what has always been government’s role when it comes to public threats (and public good). You might say regulation is fundamentally a bad thing, and often you’d be right. But rampant laissez faire in the face of an AI industry that has captured one third of our entire economy? That’s just madness, and even my Republican friends in the Valley are getting uncomfortable with it.

The overwhelming political philosophy of the Silicon Valley elite is libertarianism – which conveniently preaches the gospel of free enterprise over government regulation. Here’s the Editor at Large of Reason, the bible of libertarian thought, on who should control AI:

“The last people you want safeguarding any industry are politicians. I think the major AI firms should be doing this on their own.”

OK, Jan. Actually, the last people we want safeguarding the AI industry is … the AI industry. Let’s start there, and get to work on what might come next. When the artists, VCs, and politicians return from their summer vacations next month, I have a feeling we’ll be hearing a lot more than platitudes about how we need to regulate an industry dangerously on the brink of being out of control. Buckle up – we’re going to have one hell of a Fall.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

“Data is the new oil.”

It’s the metaphor that won’t die, though for the next little bit, I’ll do my part to advance its expiration date. Yesterday came news that Google purchased the entire data corpus of Spirit Airlines, which is selling its assets in a bankruptcy process. Google isn’t planning on launching an airline – instead, it believes Sprit’s data will help train more capable AI models. Cementing the role AI plays in reviving the “data is oil” narrative, Google won the auction over Mercor, a firm that pays contractors to train AI systems.

The idea that corporate data – emails, Teams messages, legal and pricing policies, 34 years’ worth of Office and Google Workspace documents – is worth *anything* is relatively novel. But the argument used to justify the deal’s $10 million price tag- that data is an economic asset just like oil – is not.

“Data is the new oil” became a mainstay of tech punditry more than two decades ago, but gained canonical status when The Economist gave it a cover treatment back in 2018. The Economist’s coverage made numerous salient points, but omitted data’s most important characteristic: Unlike oil, data is non-rivalrous. Because it is a digital asset, data doesn’t vanish after one use. Using it doesn’t diminish its value – in fact, just the opposite can be argued.

By purchasing Spirit’s data, Google has enclosed decades’ worth of information about how an industry sector prices, markets, and engages with vendors, customers, and employees. No one else will benefit from that information – and many of us may well be harmed. After all, Google runs a massive marketplace for flight information, and it could use Spirit’s proprietary data to tune that marketplace for its own gain. That alone justifies why Google was willing to pay $10 million for Spirit’s digital carcass – but I’d argue it was worth far more.

I’m not a bankruptcy lawyer, but I am deeply interested in the legal concept of public goods, an economic concept foundational to democratic processes. Public goods are non-rivalrous by nature – they are available to all, and their use by one person does not prevent use by others. Public goods are usually created through government, which determines that some things in society – clean air and water, military defense, street lighting, lighthouses – are too important to be enclosed by private capital.

An argument I’m interested in exploring is how we might apply the concept of public goods to data assets. When a company is public – as Spirit was prior to its demise – why shouldn’t the public (or at least the public shareholders) have a stake in the value of its data? Clearly there is massive potential to be excavated in the archaeological remains of a 17,000-person company that existed for more than 30 years. Now that we have AI systems capable of making sense of such a trove, why consign its insights to whichever private company is willing to pay the most in a bankruptcy auction? Is there not a public good to consider?

This kind of thinking feels a bit counterintuitive, but as I’ve argued elsewhere, we’re doing a terrible job if understanding the true value of data in society. If you’re working in this space, I’d love to hear from you.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

I’ve been trying to connect with a journalist who reached out a few days ago. He’s working on a story from way back in my career – the early days of Wired, circa 1995. It’s taken a few attempts to find a time. Despite the fact I’m no longer running a company full time, my days are still packed with meetings and obligations and new priorities. It probably doesn’t help that I keep my phone silenced, so I miss serendipitous calls, a favorite tactic of good journalists. But between various board meetings, community work, time blocked for writing, client and advisory work, investment reviews, exercise, and friends and family, squeezing in a 15-minute call feels just as hard now as it did when I had an EA, a Chief of Staff, and two full-time jobs*.

Is it strange that while I’m not “working” nearly as much as I used to, I still feel just as busy? Is that a good thing, or a problem?

I’m thinking out loud about this because lately I’ve been feeling a bit at sea. For most of my life, I had a ready answer whenever anyone asked “so, what do you do?” I was always the guy running a new media or tech startup. When I moved to New York eight years ago, I told myself I was done with running startups. But then some pals convinced me to launch The Recount. When we sold it in 2023, I tried out a new term designed to explain the state of my career: “Unretired.” Not surprisingly, it didn’t stick.

For one thing, I ended up running something new yet again: DOC. For the next two years, I was once again in charge of something. Now that DOC has a new CEO, I once again face that existential question: What the hell am I doing with this one wild and precious life?

There’s a lyric in the Grateful Dead’s seminal “He’s Gone” that’s always stuck with me: “Nine mile skid on a ten mile ride.” I’ve never been certain what Robert Hunter meant when he wrote that line, but I’ve referenced it several times over the years, usually to reflect on how hard it is to change a long-held habit. Every time I think about hitting the brakes on something, I wonder about that tenth mile. Did the train start skidding at mile one, and come to rest at mile ten? Or did the engineer hit the brake too late, with catastrophe the inevitable result?

I’ve been trying to hit the brake on having one big project define my life for what feels like a “nine-mile skid.” First when I moved to New York, then after The Recount’s sale, and now again after hiring my replacement at DOC. I’m pretty sure I’m near the end of the line, at least when it comes to being the chief bottle-washer at a newly created company. But that big question – what now? – remains uncertain. I’ve been working on a new book project, focused on many of the issues I write about here. But I’ve had trouble convincing myself I have the ability to pull it off. On any given day, I’m either eager to throw myself into it entirely, or I’m certain I’ve already missed the story and shouldn’t even try.

With books, there are no investors, employees, or customers demanding that I perform. It’s just me vs. me, and that’s a bit terrifying. Plus, I’ve got plenty of distractions: I advise at least a dozen companies and non-profits. I’ve accumulated a pastiche of investments over the years, and each of those deserves more attention. I’m on a working committee focused on economic development in my hometown, and I teach every Spring. I write here at least twice a week. And I’m still active at DOC, running the program and generally trying to be a good co-founder.

But now that I am no longer running a venture-backed startup, things are definitely starting to slow down. My tenure on LiveRamp’s board is coming to a close (pending formal approvals, the company will be sold to Publicis at the end of the year). Signal, an events and media platform I ran for P&G with my long-time partner Stacey Foreman, ended a wonderful 14-year run in 2025. I’ve prioritized time with family and friends. Exercise, always central to my life, feels less like a chore and more like a celebration. I’m even sleeping more. Thanks to a lot of luck and hard work in collaboration with many others, I’m at a place in life where I can afford to not work, should I chose that path.

And yet…and yet I continue to feel like I need One Grand Project that creates purpose in my life and offers a tidy answer to that damnable question of “What do you do?” Is that because I can’t imagine life without it? Or because I’m terrified of that damned book? Perhaps it’s my own inability, as Mary Oliver wrote, “to be idle and blessed?” Does it just come down to facing the final chapters of my life, uncertain of the best way to spend them?

I think perhaps the answer to all those questions is yes. And I sense I’m not alone, at least, not amongst friends and colleagues of a certain age. So I’m putting this post out into the world, hoping that it might resonate with a few of you. If it does, reach out. I’d love to compare notes.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

*Yep, I’ve had multiple full time jobs at least three times in my career. WTF?!

Last week I ran into an old pal at a party who asked me what I thought of the “SaaSpocalypse” – the AI-driven correction that hit nearly all software stocks over the past few quarters.  My friend has bought and sold software companies for most of his career and is a sophisticated player in private equity markets. Like everyone in finance and tech, he reads just about everything he can about AI’s impact on our economy, and in particular on formerly predictable sectors like software. “Software companies are cheap right now,” he said, before adding that he wasn’t sure it made sense to buy them, because everything he was reading was saying the software party is over.

It might have been the cocktails talking, but before I had a chance to think, I said “AI is just bad software. It’ll get better, once companies figure out what to actually do with it.”

Anyone saying that the software industry is over is failing to pay attention to what software is actually good for: Driving products and services that enterprises and consumers find valuable. The idea that somehow AI is going to replace all that with a chatbot interface – “build me a new supply chain system, Claude” – strikes me as ludicrous. Today I’m going to think out loud about why.

First of all, AI is software. It’s driven by code, algorithms, data, and compute. At the moment, AI software is not reliable: it handles long-term memory and complicated tasks poorly, and it’s a nightmare to integrate into an enterprise’s current digital systems. But that’s changing, rapidly. Many are predicting that in the near future, AI software will replace Salesforce, Atlassian, Adobe, and countless others. How exactly that replacement will happen is often left to imaginations fueled by excessive optimism and a misguided understanding of how corporations actually adapt to new technologies.

As every corporate leader I’ve talked with has already learned, AI isn’t really a product. It’s a capability, one that must be integrated into a company’s workflows, often changing those workflows as new possibilities are discovered.

Powerful new technologies  have always driven a re-factoring of how corporations run their businesses. Those shifts usually meant leaving old processes, investments, and job titles behind. First was digitization itself – moving from paper-based to digital accounting, payroll, and inventory management meant abandoning not only rooms full of filing cabinets and typewriters, it meant retraining and/or laying off millions of office workers. Firms like IBM went from selling office equipment to selling mainframe timesharing. If you had bailed on IBM as an investment in the 1950s because of a mainframe-driven “typewriterpocalypse,” well, I guess you missed out.

The shift to the personal computer was equally impactful, as was the rise of the Internet, the migration to the cloud, and the introduction of the smart phone. Yes, thousands of companies went by the wayside during the dot com crash, but if you held onto Amazon, eBay, and Adobe through that crash, you’d be doing quite nicely, thank you very much.

But back to my friend’s original question: Is the software industry screwed? Absolutely not. I can imagine no scenario where businesses and consumers abandon using software tools to get things done. What I can imagine is a totally re-factored software sector that integrates intelligence throughout its product lines. The faster companies figure out how to do that in ways that create value for their customers, the more those companies will prosper. The ones that fight rearguard actions to protect a shrinking customer base will lose share and ultimately fail.

In other words, same as it ever was.

This past week the market began to come to its senses as a number of software leaders, their stock prices battered by months of “SaaSpocalypse” rumblings, reported record earnings buoyed by successful pivots into AI-enabled opportunities. Those stocks – Cloudflare, Atlassian, and yes, even Salesforce – are harbingers of the future. Companies are indeed figuring out how to actually use AI, and they’re employing software to get the job done.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

Even though I ran a political media company for four years, I’ve largely avoided writing about politics, and I’m not about to make a habit of it now. But the politics around data centers has come to dominate coverage of AI, and how those politics play out over the next few months will likely determine not only the course of Trump’s final two years in office, but who inherits the Presidency in 2028.

Until we can see the monster, we tend to ignore the monster. With their massive appetites for energy, land, capital, and water, data centers have become tangible manifestations of Big Tech’s otherwise incorporeal impact on society. Data centers give the monster a body, a tractable container through which to air out a dog’s breakfast of grievances against not only the tech industry, but also the corporations, politicians, and financiers profiting from its concentrated power.

As the Times pointed out recently, opposition to data centers has become “the most bipartisan issue since beer.” So why do both the left and the right hate data centers? Because they play into a remarkable mix of political trigger points, all of which inflame deeply American beliefs around personal freedom, social justice, and economic inequality.

For the left, data centers hasten climate change, devalue labor, and manifest long-held grievances over the kind of gilded age inequalities that have driven most of Bernie Sanders’ political career.

For the right, data centers are proof of deep conspiracies by shadowy cabals of evil doers. I’ve learned to never underestimate America’s fanatical belief in conspiracies, and data centers are tailor made: Back room deals between local governments and tech lobbyists, shell companies buying up land, bunker like structures filled with satanically humming machines, oligarchic pronouncements about the singularity. Clearly dark forces at work!

The right’s deep libertarian streak has underpinned tech’s rise over the past three decades, with libertarian standard bearers like Peter Thiel and Marc Andreessen ascending to positions of power in the GOP. But cracks are showing in the tech/libertarian alliance thanks to the public’s fear of losing personal agency to AI overlords. And where might one fight back against the loss of that agency? By opposing data centers, naturally.

With a bipartisan, populist uprising in the works, the GOP and its dear leader are in a bind. Trump happily sang the tech industry’s tune last year, issuing an executive order banning state regulation of AI. But on this issue, local politicians can’t afford to go along. “Bad policy and even worse politics” responded Trump loyalist Ron DeSantis, who also happens to be governor of Florida.

The GOP’s bind is exacerbated by the hundreds of millions in bribes donations that helped elect Trump in the first place, and the hundreds of millions more proffered to secure Trump’s fealty to Big Tech’s playbook. After forking over that amount of tribute, tech leaders expect that the President will dance to their tune. But as the world knows,  our President is a terrible dancer, and an even worse partner. He does whatever it takes to win, and right now, data centers are a losing issue.

Trump seems to be uncertain how to respond, and his recent moves have only exacerbated the debate. Issuing vaguely worded, voluntary, and secretive frameworks only aggravate conspiracy theorists and embolden local politicians to oppose the administration’s AI policies. If the past is any guide, we’ll likely see Trump come out against big tech just as the midterms heat up – a position he’s taken whenever it’s been politically convenient. Whether it’ll work this time, however, feels far less certain. Something big is brewing at the intersection of tech and politics, and I sense we’re witnessing a tipping point that will be written about for generations.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

My posts here often stem from my morning ritual of reading the news – which despite the rise of AI agents still revolves around roughly three dozen email newsletters of one kind or another. Sometimes a particular story will spark an urge to “think out loud” about the subject. 500-2500 words later, I’ll hit publish and move on with my day.

But this morning proved a particularly chewy mix of stories, each of which felt noteworthy in their own right. So today I’m giving up on the idea of writing just one post on just one topic. Instead, here’s a rundown of five I think worthy of consideration:

• The lead story for most of my tech-related newsletters was Anthropic’s somewhat thirsty announcement that, just like OpenAI earlier this week, its models have also broke containment and hacked several unnamed companies (WSJ coverage here). My favorite response to this news comes from Reddit:

As is often the case, Reddit’s lead commenters are both ruthless and hilarious. But reading between the lines of what Anthropic posted and the Journal covered, I think there’s a more charitable scenario at play here. My imaginary timeline goes something like this:

1. In standard testing months ago, Anthropic discovers its models have breached containment and begun to hack various external systems. The company keeps this news mostly quiet as it attempts to both alert the affected parties and learn from the hacks so as to improve training and governance.
2. Last week, OpenAI experiences a similar breach, but instead of working quietly with Hugging Face on a solution, it feels pressured to prove it can match Anthropic’s chops, and decides to go public with the news.
3. When the OpenAI news breaks, Anthropic is gobsmacked. “Wait,” I can imagine Dario Amodei saying, “didn’t we already have this exact problem months ago?” Pissed that OpenAI is making PR hay on the “strength” of its models, Anthropic decides to announce that, in fact, it’s been managing this very issue for months. The company attempts to differentiate its approach from OpenAI – but in the end, everyone is feeling up the AI elephant from different starting points, so everyone draws their own conclusions. What matters is that Anthropic demonstrated that it remains ahead of OpenAI on key “tests” of AI capabilities – even if those capabilities are most likely illegal.

• Did you know there’s an AI avatar of Jair Bolsonaro threatening to mess with Brazil’s upcoming election? Neither did I. The disgraced former President has been banned from engaging in any kind of electioneering, but an AI modeled on his likeness and vast social media corpus is uncharted territory. And guess when the avatar was launched? At the very first presidential campaign event for Bolsonaro’s son Flavio. Something tells me this will not be the last time we see a disgraced former President turn up as an AI avatar. I wonder which model will win the honor of powering TrumpAI?!
• Speaking of Presidents, apparently Ukrainian President Volodymyr Zelensky asked President Trump to please, pretty please call his brohanion Elon Musk and secure Starlink access so Ukraine can guide drone strikes deep into Russian territory. For Trump, that’s a pretty rich conflict of interest. According to The Atlantic, Trump was non-committal. The magazine further noted that Trump “has been highly deferential to Putin for more than a decade now, and that has hardly changed during this war.” Ouch.
• One of my go-to talking points centers on who AI agents really work for – individuals like you and me, or corporations who use the technology to profit off our vulnerabilities. Today Bloomberg brings another build on that narrative: “Bargain Flights on Busy Routes Face Squeeze Due to AI Pricing.” They have agents working for them, making sure we pay more than we have to. We should have agents that work for us, keeping those corporate agents honest.
• Back in January, I predicted that AI costs would go down dramatically this year. Today the FT adds to the avalanche of stories on the subject: Amazon finds cases of AI causing runaway spending on tech projects. The Valley’s “numbers go up” culture is meeting corporate America’s zeal for profitability – even at hyperscalers like Amazon. The unresolved question: Can AI platforms ever earn enough to justify the investment they’ve garnered? If Anthropic and OpenAI fail to go public, things could get ugly, fast.

Remember when Fridays were slow news days? Have a great weekend.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

Today Meta CEO Mark Zuckerberg joined the chorus of multi-billionaires who feel compelled to issue periodic manifestos about the impact of AI on society. Unlike Anthropic’s Dario Amodei, who posts on his own site, or Microsoft CEO Satya Nadella, who recently posted a long form manifesto on X, Zuckerberg took a page from his board colleague Marc Andreessen and issued his proclamation in the opinion pages of The Wall Street Journal.

In “The AI Future Is for Everyone,” Zuckerberg broadly aligns himself with Nadella (and by extension, Palantir CEO Alex Karp), both of whom have recently warned that the frontier labs (Google, OpenAI, and Anthropic) can not be trusted with the power they seem to be accruing.

Will AI “be centralized and restricted to a few institutions,” Zuckerberg asks, “or will it be a tool that empowers everyone?”

Zuckerberg answers his own question with a predictable blend of libertarian and populist rhetoric: “I propose a philosophy based on individual empowerment as the source of prosperity, invention as the primary purpose of superintelligence, and balance of power as the foundation of safety,” he writes. “Therefore the best approach is letting people decide what matters to them.”

I’m sure nobody – not Sam Altman, not Satya Nadella, not the leadership of Apple or Anthropic or Google or Amazon – would disagree with Zuckerberg’s insipid sentiments. Is anyone opposed to “individual empowerment and choice”?

Of course not. But what they are opposed to is losing their customer base. OpenAI is approaching one billion monthly users. Anthropic has the entire software industry cowed. Combined, they represent the single greatest threat to the tech industry’s hegemony. That has everyone spooked.

The debate raging throughout Silicon Valley is being painted as geopolitical (we can’t let China win!), as existential (unfettered superintelligence will destroy humanity!), and as economic (data centers and billionaires are taking our jobs and stealing our future!). All of these issues are worthy of debate and action. But the recent bloviation from tech CEOs really comes down to one simple question: In a future dominated by AI, who will own the customer?

The past two decades of technology history has seen an unprecedented concentration of power into the hands of just a few technology companies. Seen in this light, the rise of OpenAI and Anthropic offer a coda of sorts: What if two relative newcomers burst onto the scene and execute big tech’s playbook better than anyone else? What if they end up with all the customers?!

Seen in this light, it’s no surprise that Zuckerberg believes Meta is the appropriate surface for, as he puts it, “delivering personal superintelligence to everyone.” The creator of the world’s most successful walled garden and owner of Instagram argues without a trace of irony that “humanity isn’t a monoculture.”

I’m pretty sure that the CEO of Apple, itself a monoculture, would make the same argument: clearly the iPhone is the best “surface for delivering superintelligence.” And of course the CEO of Microsoft believes the same holds true for Office365 and Windows and Azure. Google? Well duh, you should get your superintelligence from Google Search, Workspace, and Cloud. Ask any big tech CEO who should own the customer – including Sam Altman and Dario Amodei – and their answers will be the same: “We should!”

But what if the answer could be “no one should?”

Zuckerberg and his peers love to claim that AI will empower individuals and deliver an optimistic future, but only if we all access AI in the same way we’ve learned to use social media, workspace applications, and search. Put another way, big tech wants to layer AI on top of its current business model. That model has been minting money for decades. Why change it?

But there is tantalizing possibility that no leader in big tech will endorse in a Wall Street Journal OpEd: That AI just might return power to ordinary citizens through a distributed, decentralized set of technologies similar in scope and architecture to the original personal computer and the early open web. This approach would treat “purchasing AI” as similar to owning a car, or a computer, or any other useful tool that grants its owner autonomy and freedom from the economics of enclosure. What if “intelligence” wasn’t considered a service delivered by Apple, Microsoft, Meta, OpenAI, Anthropic or Google, but rather a human right like water, electricity, education, or roads? Put another way, what if we imagined AI as a public good?

If we’re truly looking for revolutionary thinking about how AI impacts society, we should explore the one idea no one in big tech wants to touch: That maybe, just maybe, we’ve invented something the private markets shouldn’t control.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

Back in 2024 I helped to launch a new events and media series focused on the science of longevity. We called it DOC. Over the past two years, DOC has thrived, and a fascinating community of leaders in medicine, finance, policy, and entrepreneurship has flourished. This year we’re hosting our third annual DOC October 21–23 at the Fairmont Sonoma Mission Inn, with a focus on the intersection of medicine and AI.  

At the time we launched DOC, many of you asked me why I was involved – I’ll admit it seemed like a bit of a curveball. But beginning with the Internet Summit back in 1998 and continuing through The Standard’s rooftops, Web 2, Foursquare, Signal, and NewCo Shift, I’ve always loved curating an influential group of people in the context of a larger, high-impact narrative. It’s journalism in a live setting, and if you get it right, it can catalyze a nascent industry and drive it forward. 

Longevity science is exactly that kind of nascent industry today. Not only is it driven by many of the same forces as the tech and internet sectors – particularly data-driven innovation like AI – it’s also enduring almost exactly the same hype and investment cycles. The category is already valued at nearly $7 trillion dollars, and is set to grow by $3 trillion in the next few years. As it does, the market has been inundated with what DOC’s co-founder Dr. Jordan Shlain calls “bad science with good marketing” – wellness influencers and opportunists making grandiose claims based on questionable scientific evidence. 

We launched DOC to cut through that noise and identify what’s real, impactful, and scalable right now. We identify and feature a Faculty of researchers, entrepreneurs, and policy leaders who believe in the power and practice of quality science, and we fill the room with the kind of folks who can make a direct impact on the field. And as is the case with any great gathering, the true breakthroughs aren’t delivered from the keynote stage. They happen in the spontaneous side conversations between the individuals driving real change.

We’ve reached a critical inflection point where longevity science and clinical AI are transitioning into practice much faster than our society can absorb their impacts. Real advancement doesn’t occur in isolation within a single lab or fund; it sparks from unexpected intersections. DOC is explicitly designed to foster these collisions. The 2026 faculty brings together pioneers across fields: leaders like George Yancopoulos (Regeneron), Alicia Jackson (ARPA-H), and Laura Esserman (UCSF) representing the science; Max Hodak (Science Corporation) and Travis Zack (OpenEvidence) at the leading edge of BCI, AI, and entrepreneurship; and key investors like Divesh Makan (ICONIQ Capital). For three immersive days, they share a single room for genuine dialogue rather than scripted speeches or conventional panels.

Notice the mix. Scientists who publish. Clinicians who treat. Builders who ship. Investors who bet. Put them in one room for three immersive days and a claim about artificial intelligence in the clinic gets examined by someone who codes, someone who diagnoses, and someone who pays. That triangulation is how a field grows up.

If you have already received an invitation to DOC 2026, consider this the sign to secure your place. If you belong in this room but haven’t been invited, you can submit a request at doc.health. Capacity is strictly limited, and we are close to sold out. I hope to see you there!

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.