B (8:16)

Again, what's going on over at Apple? He since walked this back with a statement to Apple employees, but reporting earlier suggested Apple's chip chief Johnny Sruji told Apple CEO Tim Cook that he is seriously considering leaving Apple soon. Quote Sruji, the architect of Apple's prized in house chips effort, has informed colleagues that he intends to join another company if he ultimately departs. It all adds up to one of the most tumultuous stretches of Tim Cook's tenure as CEO. Though Cook himself is unlikely to leave imminently, the company has to rebuild its ranks and figure out how to thrive in the AI era within the company. Some of the departures are cause for deep concern, with Cook looking to stave off more with stronger compensation packages for key talent. In other cases, the exits just reflect the fact that veteran executives are nearing retirement age. Still, many of the shifts constitute a disconcerting brain drain. While Cook maintains that Apple is working on the most innovative product lineup in its history, a slate that's expected to include foldable iPhones and iPads, smart glasses and robots, Apple hasn't launched a successful new product category in a decade that leaves it vulnerable to poaching from a range of nimbler rivals better equipped to develop the next generation of devices around AI. The flurry of retirements reflects a demographic reality for Apple. Many of its most senior executives have been at the company for decades and are roughly the same age, either in their 60s or nearing it. Cook turned 65 last month, fueling speculation he would join the exodus. People close to the executive have said that he's unlikely to leave soon, though succession planning has been underway for years. John Ternus, Apple's 50 year old hardware engineering chief, is considered by employees to be the frontrunner CEO candidate. When Cook does step down, he's likely to shift into the chairman job and maintain a high level of influence over the iPhone maker. That makes it unlikely that Apple will select an outsider as the next CEO, even as executives like Nest Labs founder Tony Fadell are being pushed as candidates by people outside the company. Though Fadell helped invent Apple's iconic iPod, he left the tech giant 15 years ago on less than friendly terms. End quote. I'm not as much of an Apple Kremlinologist as some, but my original instinct about all this was people inside Apple knew the Tim Cook regime was coming to an end sooner rather than later, and for whatever reason wanted to head for the exits before others did. But now, that definitely doesn't look like it can happen, right? Cook can't leave Hot on the heels of these other departures.

B (10:49)

We all remember the choices that shaped the course of our lives in business. World renowned venture capital firm Sequoia Capital calls them Crucible Moments. Their podcast brings you inside the pivotal decisions that defined some of today's most influential companies. Hosted by Sequoia's Rudolf Botha, Crucible Moments Season 3 pulls back the curtain on the untold stories behind companies like Stripe, Zipline, Palo Alto Networks, Klarna Supercell, and more. I loved the recent episode with the founder of Zipline and how even late to the game, they are leapfrogging the bigger players to bring true autonomous drone delivery not just to hospitals, but now to customers of the likes of Walmart and Chipotle. I might win that burrito delivery bet someday soon. Tune in to Sequoia's new season of Crucible Moments to discover how some of the most transformational companies of the modern era were buil. Crucible Moments is available everywhere you get your podcasts and@CrucibleMoments.com.

B (11:44)

Keeping pace with data growth in the age of AI is like trying to find enough shelf space after a trip to a big box store. AI and data growth have outpaced the old storage model. Manual management of traditional storage can't keep up, so it's time for a new, unified approach from Pure Storage. They help organizations simplify and automate how data is stored and managed, eliminating silos and putting intelligence at the center of operations. When you don't know where data lives or how it's used, governance slips, visibility and compliance can become constant challenges. The Pure Storage platform unifies storage into a single intelligent layer that can turn data into a governed, virtualized cloud of data with guaranteed outcomes. Learn more@PureStorage.com Morning Brew that's PureStorage.com Morningbrew.

B (12:37)

Meta has acquired Limitless, which makes a pendant style AI wearable that records and transcribes real world conversations. Limitless had raised more than $33 million. Quoting Reuters earlier this week, Meta hired longtime Apple executive Alan Dye, a move widely seen as sharpening its focus on next generation devices. Meta plans to use Limitless technical capabilities as part of its development of next generation AI enabled wearables. Meta currently has partnerships with Essilor, luxottica Brands, Ray Ban and Oakley to make AI powered smart glasses. We're excited that Limitless will be joining Meta to help accelerate our work to build AI enabled wearables, a Meta spokesperson told Reuters. Limitless, formerly known as Rewind, makes a wearable pendant that clips to clothing or a lanyard. The device records conversations and can generate transcripts and produce searchable summaries through a companion app. It is part of a growing category of AI assistants designed to augment memory and everyday productivity. Meta and Limitless will continue supporting existing users, but the company will stop selling devices to new customers. Limitless said existing users will be asked to accept revised privacy terms to maintain service. Limitless has raised more than $33 million from investors, including Sam Altman and A16Z end quote.

B (13:56)

Another potential IPO for next year that would be like the biggest of all time. Sources say SpaceX is planning a secondary share sale that value the company at $800 billion, surpassing OpenAI to make SpaceX the most valuable US private company. Quoting the Journal, the company's chief financial officer, Bret Johnson told investors about the sale in recent days. And SpaceX executives have also said the company is weighing a potential initial public offering in 2026. Some of the people said there is no guarantee SpaceX will reach the $800 billion valuation it is aiming for, though the company has a dedicated investor base that routinely writes checks for ever growing valuations. Elon Musk said on X in June that the Texas expected to generate about $15.5 billion in revenue for the year. While much of Elon Musk's business empire is facing growing challenges, his rocket and satellite company remains stronger than ever, thanks in part to its dominant position launching rockets into space. Many investors say the company's satellite business Starlink, which has more than 8 million active customers, is also driving up its big valuation. End quote an analysis of more than 100 trillion tokens from the past year shows how AI usage is evolving. For example, reasoning models now represent over half of all AI usage, and open weight model use has grown steadily. Quoting OpenRouter, AI programming has become the most consistently expanding category across all models. The share of programming related requests has grown steadily through 2025, paralleling the rise of LLM assisted development environments and tool integrations. Programming queries accounted for roughly 11% of total token volume in early 2025 and exceeded 50% in recent weeks. This trend reflects a shift from exploratory or conversational use toward applied tasks such as code generation, debugging and Data scripting. As LLMs become embedded in developer workflows, their role as programming tools is being normalized. This evolution has implications for model development, including increased emphasis on code centric training data, improved reasoning depth for multi step programming t and tighter feedback loops between models and integrated development environments. Among the highest volume categories, roleplay stands out for its consistency and specialization. Nearly 60% of roleplay tokens fall under games roleplaying games, suggesting that users treat LLMs less as casual chatbots and more as structured role playing or character engines. This is further reinforced by the presence of writers resources at 15.6% and adult content at 15.4%, pointing to a blend of interactive fiction, scenario generation and personal fantasy. Contrary to assumptions that roleplay is mostly informal dialogue, the data show a well defined and replicable genre based use case. Programming is similarly skewed with over two thirds of traffic labeled as programming other. This signals the broad and general purpose nature of code related prompts. Users are not narrowly focused on specific tools or languages, but are asking LLMs for everything from logic debugging to script drafting. That said, development tools at 26.4% and small shares from scripting languages indicate emerging specialization. This fragmentation highlights an opportunity for model builders to improve tagging or training around structured programming workflows. By contrast, finance, academia and legal are much more diffuse. Finance spreads its volume across foreign exchange, socially responsible investing and audit accounting. No single tag breaks 20%. Legal shows similar entropy, with usage split between government other at 43% and legal other at 17.8%. This fragmentation may reflect the complexity of these domains or simply the lack of targeted LLM workflows for them. Compared to the mature categories like coding and chat, the data suggests that real world LLM usage is not uniformly exploratory. It clusters tightly around a small set of repeatable high volume tasks. Roleplay programming and personal assistance each exhibit clear structure and dominant tags. Science, health and legal domains, by contrast, are more diffuse and likely under optimized. These internal distributions can guide model design, domain specific fine tuning, and application level interfaces, particularly in tailoring LLMs to user goals. The share of total tokens routed through reasoning optimized models climbed sharply in 2025. What was effectively a negligible slice of Usage in early Q1 now exceeds 50%. This shift reflects both sides of the market. On the supply side, the release of higher capability systems like GPT5, Claude 4.5 and Gemini 3 expanded what users could expect from stepwise reasoning. On the demand side, users increasingly prefer models that can manage task state, follow multi step logic and support agent style workflows rather than simply generate text. A central question in the AI ecosystem is the balance between open weight that we abbreviate to OSS for simplicity and proprietary models. The figures below illustrate how this balance has evolved on OpenRouter over the past year. While proprietary models, especially those from major North American providers, still serve the majority of tokens, OSS models have grown steadily, reaching approximately one third of Usage by late 2025. A significant share of this growth has come from Chinese developed models. Starting from a negligible base in late 2024, weekly share as low as 1.2%, Chinese OSS models steadily gained traction, reaching nearly 30% of total usage among all models in some weeks. Over the one year window, they averaged approximately 13% of weekly token volume, with strong growth concentrated in the second half of 2025. For comparison, row OSS models averaged 13.7%, while proprietary row models retained the largest share, 70% on average. The expansion of Chinese OSS reflects not only competitive quality, but rapid iteration and dense release cycles. Models like Quen and Deepseek maintain regular model releases that enabled fast adoption to emerging workloads. This pattern has materially reshaped the open source segment and progressed global competition across the LLM landscape.

B (20:10)

You know what I did last night? For the first time ever, curling. Prospect park has an ice rink in the winter and they have Sunday night curling league. So I did a total rookie session with about a dozen other people and man it was great fun. Would definitely recommend trying curling. If anyone listening to me right now is in that Brooklyn Curling league, get in touch because I might want to pursue this further. Talk to you tomorrow.

A (20:41)

The Uniswap Wallet makes it easier and safer to own and use crypto Created by pioneers of the crypto economy, the Uniswap protocol has powered over $3 trillion in trading volume and it's trusted by tens of millions worldwide. With the Uniswap Wallet, you can discover, swap and manage your crypto all from your phone. Buy your first crypto assets in just a few taps and start exploring the freedom of decentralized finance. With Uniswap Tap the banner to get started.