2025-04-14 21:19:13
Hi all,
Here’s your Monday round-up of data driving conversations this week — all in less than 250 words.
Today’s edition is brought to you by Attio, the CRM for the AI era.
Sync your emails and Attio instantly builds your CRM with all contacts, companies, and interactions enriched with actionable data.
Trade tensions. China suspended exports of key rare earth minerals to underscore its control over roughly 90% of global supply of 17 strategic elements.
Ukraine’s drone advantage. Ukraine now manufactures FPV drones domestically at a lower cost than Chinese imports.
OpenAI’s dominance. 10% of the world now uses the firm’s AI, according to Sam Altman and monthly revenue is up 30% in the last three months at $415 million.
Academia embraces AI.
Breaking Wiki’s bank. AI data scraping increased Wikipedia’s infrastructure costs by 50% since January 2024.
Solar surge continues. Solar marked its 20th year as the fastest-growing power source in 2024.
The Detroit Three. Three major US car manufacturers’ global market share fell from 29% to ~13% since the early 2000s.
The American dream. Who’s going to do it?
Thanks for reading!
Today’s edition is brought to you by Attio
Attio is the AI native CRM built for the next era of companies.
Sync your emails and watch Attio build your CRM instantly - with every company, contact, and interaction you’ve ever had, enriched and organized. Join industry leaders like Flatfile, Replicate, Modal and more.
2025-04-13 08:44:43
Hi, it’s Azeem with our weekend email which will give you the much-needed distance from the headlines. Google is going from strength to strength, releasing its new Ironwood TPU, while a looming economic squeeze might accelerate AI adoption in ways we’ve seen before with past technologies. And all the while, individuals seem to be racing ahead in using AI tools day-to-day, even as big organizations struggle to keep up. Let’s dive in.
Two years ago, I warned that Google could be facing a tidal wave like never before. I wrote:
It can be hard for successful firms like Google to address threats like [OpenAI]. The firm has some of the best research, development and engineering in this field. So too did many firms facing competitive pressure before it. Digital Equipment led in minicomputers as IBM launched the PC. Microsoft had tons of online and Internet know-how. Nokia knew more about smartphone operating systems than either Apple or Android. [...] If the start page of the Internet is shifting further from the browser and Google.com (or its preferential position on iPhones), that is more of a challenge. [...] Microsoft, under Bill Gates, survived the Internet tidal wave. Google, too, could prosper during the GPT tidal wave. But there is a lot to play for.
This week, Google has shown us that it’s clearly pulling ahead in terms of model performance. The chart below shows the Pareto frontier of LLM models – that is, for every combination of price vs performance, which model performs the best. Google’s family of Gemini models trounce everyone else… for now. Undoubtedly, there is some good research driving these gains, but is Google’s secret weapon its decade-long investment in custom hardware?
Undoubtedly, there is some good research driving these gains, but is Google’s secret weapon its decade-long investment in custom hardware?
2025-04-12 21:28:58
ChatGPT now has one billion weekly users.
The news came out through a tweet by investor Steve Jurvetson who was attending the TED Conference in Canada yesterday, with my emphasis added.
Sam Altman: “Last we disclosed, we have 500 million weekly active users, growing fast.”
Chris Anderson: “But backstage, you told me that it doubled in just a few weeks.”
So, one billion weekly users is the private number, and likely the real one. In his on-stage discussion with Chris, Sam also alluded to 10% of humanity (call that 800m) using ChatGPT.
Both of these are substantially up from the 400m weekly number OpenAI’s Brad Lightcap spoke about three weeks ago. At a minimum, it is that doubling in just a few weeks—unprecedented at this scale.
Let’s get this straight: that is an enormous number of users. These are weekly active users, a key measure of retention and habituation.
This isn’t the peak.
Weekly users become daily users. Daily users realize what AI can do for them. Personal usage skyrockets as you shift Google searches to AI, as you test all your ideas and thinking—from sales emails to shopping lists—against an AI, as you call on coding assistants to build mockups or working apps.
Each of those applications demands more AI, more tokens, and more compute. The hockey stick goes exponential.
As a new dad, Sam also reflected that his “kids will never be smarter than AI.”
My takeaways:
This is a psychological tipping point. A billion of anything is a lot.
Competition will heat up. Google (as I’ll write tomorrow) has better models across the board. OpenAI has better products. Both firms, and their big competitors, will figure out how to go even faster.
Prepare for a greater surge in usage.
The way we interact with technology is going to change over a period measured in 2-3 years rather than five or more.
Cheers,
Azeem
P.S. ’s full tweet is reproduced below.
2025-04-12 17:42:56
Globalization has grown exponentially. Our linear institutions, norms, values and systems could not adequately keep up.
It is a case of what I call the “exponential gap”, the chasm that emerges between rapidly developing new technologies and the slower rate at which our existing corporations, institutions, employees, politics, and wider social norms adapt to these changes. In this case, the bundle of technologies, some digital, physical, social and legal technologies, is called “globalisation”.
This set of technologies refactors how the world operates faster than our existing institutions can adapt. At some point, something would have to give.
The last 30 years have since two parties spin the flywheel of globalisation faster. On the one, the US consumer, who has kept spending, is supported by common-sense decisions by businesses to reduce their costs. However, those short-term decisions have turned into long-term structural problems. Like plaque on teeth, they are layered and layered.
China is the best place to make an iPhone and forget the cost. The know-how, over 20 years, is there. The ecosystem, with its codes, cues and knowledge, is there. Hear Tim Cook on this. Mark Gurman goes into much more detail. Moving advanced manufacturing of scale takes time, longer than a Presidential term.1
The other party, the Chinese, turned their economy into a supply-side juggernaut. It was fuelled not by domestic consumption, even though there was an enormous number of increasingly rich consumers, but through exporting to the rest of the world. The Chinese consumer doesn’t. And their relative unwillingness to spend has been recognised as a problem for a while.
2025-04-12 01:22:20
This week, I sat down with , CEO of Flexport, at a moment of real upheaval in the global trading system. Ryan runs a logistics platform that sees about 1% of all US trade, so he’s got a direct line into how companies are reacting in real time: pausing shipments, renegotiating supply chains and in some cases, facing existential choices.
I wanted to understand how the tariff storm is playing out on the ground—and what it means for the future of globalization.
We talked about:
The economic shock hitting importers right now and why many are considering abandoning shipments altogether
How this new trade war may unintentionally increase global shipping, as supply chains stretch and reroute
The weird inefficiencies emerging around semiconductors and AI hardware, thanks to tariff classifications
Whether this is just another blip—or a real inflection point in the 800-year rise of global trade
What Flexport has seen in customer behavior—and what that tells us about resilience, agility and Darwinian survival.
We also went deep into the geopolitics of shipping: the fragility of sea lanes like the Red Sea, the underinvestment in US ports, and what it really means when the US Navy stops guaranteeing free trade routes.
Ryan brought a perspective that’s part historian, part operator and part futurist. If you’ve ever wanted to understand how technology, tariffs, and tectonic shifts in globalization collide - this is your episode.
Thank you , , , , and many others for tuning into my live video with !
2025-04-09 23:13:39
Last week, I wrote to my portfolio companies with a simple message: the game is changing fast. The global order is being renegotiated in real time, driven by the inversion of globalization and the rise of AI.
These forces are creating confusion, hesitation and for the sharpest operators – opportunity.
If you’re a founder, executive or investor, this message is for you.
Sent on 6 April 2025
We are living through an incredibly weird time. Two massive waves are crashing down upon us: the accelerating force of artificial intelligence and the dramatic inversion of geoeconomics. Each of these forces is disruptive on its own – but together, they are increasing the stakes for everyone.
In this message, I will address:
The geoeconomics wave
The AI wave
My advice to founders and business leaders
Let’s start with geoeconomics and the shifting posture of the global order.
Globalization, as we’ve known it for the past three decades, is being renegotiated, restructured from the ground up. This isn’t just a US phenomenon. Governments around the world are adapting to this new paradigm, where industrial policy, national resilience and sovereign control take precedence over global efficiency.
As political winds shift and global instability intensifies, so too does the appetite for risk and experimentation. Uncertainty is now the defining condition.
