Tech Brew Ride Home - Who Inherits Anthropic?

Episode Date: October 27, 2026

OpenAI claimed three fresh breakthroughs in math, physics, and computer science — including that the Exponential Time Hypothesis is false. Apple unveiled the M6 MacBook Pro. Microsoft offered buyouts to ~60K workers, a quarter of headcount. Plus, Anthropic S-1 nuggets. OpenAI announces three breakthroughs in computer science, math, and physics, including that the Exponential Time Hypothesis, a longstanding conjecture, is false (OpenAI) OpenAI used external “shepherd” experts to guide its computer science and physics results, while its AGMAI panel of math advisers recommends AI companies drop the practice, citing unequal access, borrowed reputation, and biased selection (The Verge) Apple unveils the 14-inch MacBook Pro with M6, with a 12-core CPU and a 12-core GPU, up to 32GB of unified memory, and Wi-Fi 7, from $1,999 on November 6 (Apple) The M6 MacBook Pro, the new base model replacing the M5, charges to 50% in 30 minutes, adds Bluetooth 6, and opens for pre-order tomorrow, October 28, in silver and space black (9to5Mac) Microsoft offers buyouts to ~60K workers, including at a VP level and below across the US, Canada, India, and the UK; its headcount at the end of June was 223K (Bloomberg) S-1: as of September 30, Anthropic had 11,900+ customers spending $100,000+ annually, 1,900+ customers spending $1M+, and ~250 customers spending $10M+ (TechCrunch) Anthropic’s IPO filing contains an unusual succession plan: each co-founder can name a confidential ranked list of up to 10 successors to their voting power, subject to approval from five of the other six co-founders (Forbes) Anthropic’s 85% gross margin on serving existing models and $3.5B in free cash flow on $16.3B of first-half revenue is a reality check for skeptics of pay-as-you-go AI economics (The Information) Subscribe to the ad-free feed.

Welcome to the TechBrew right home for Tuesday, October 27th, 2026. I'm Brian McCullough today. OpenAI claimed three fresh breakthroughs in math, physics, and computer science, including that the exponential time hypothesis is false. Apple unveiled the M6 MacBook Pro. Microsoft offered buyouts to around 60,000 workers at quarter of their headcount, plus Anthropic S1 Nuggets. Here's what you miss today in the world of tech. AI has changed your entire business, but your cybersecurity didn't get the memo. No worries, Rubrik can help your cybersecurity catch up. With agentic cyber resilience, Rubrik built a platform that advances AI at your organization without sacrificing security. That's one platform created to handle all of your data, identities, and AI systems, so nothing gets missed or overlooked. Rubrik keeps your teams ready for whatever goes sideways with agentic cyber resilience, so your leaders have, one less reason to worry about AI usage. Agentic cyber resilience even automatically detects and responds on its own without needing a human to catch it first. Stay moving with AI. Learn how to embrace AI without putting your security at risk at rubric.com slash mb. That's rubr-i-k.com slash mb.

OpenAI has announced three new breakthroughs in computer science, math, and physics, including that the exponential time hypothesis, a. long-standing conjecture is false, quoting the verge. Less than a month after its mass release of hundreds of manuscripts that it said contained solutions to long-standing mathematics questions. OpenAI is back with three more major discoveries in math, computer science, and physics. What's different this time is the choice to have external experts that Open AI describes as shepherds guiding the results in computer science and physics, while for the math problem,

it says that they advised it to release the AI output for researchers to work through their own explanations. Opening eye added the results to GitHub today, quote, local smoothing for the Euclidean wave equation. The proof uses an auxiliary partial differential equation to establish sagis local smoothing for waves in every spatial dimension at least two. It also yields the Kakea maximal and dimension conjectures

and Stein's spherical restriction estimates through interior of the conjectured range. We are releasing the model output directly as we did in our October 6, The exponential time hypothesis is false. The paper gives a deterministic algorithm that exactly counts satisfying assignments of fixed-width, CNF formulas in sub-exponential time and polynomial space. The case of three-sat refutes the exponential time hypothesis. An independent working group of computer scientists will examine and explain the proof at the group's request. We are making the materials public immediately so others can examine them as well. This approach could have its own issue. issues, the company admits, and the AGMAI panel of math advisors OpenAI put together just said yesterday that we recommend that AI companies do not adopt the practices of designating certain people to receive the output first, citing unequal access, borrowed reputation, and bias selection as a few of the

issues. Instead, they recommended that in the future AI labs release output on scholarly repositories with lean formalized proofs, info on how the problems were chosen, funding for people working on understanding the results and other information, end quote, and quoting Engadget. The exponential time hypothesis is an assumption in computer science that some problems are just too hard to solve quickly. As Hove and A.I. notes, ETH is widely believed to be true and has been the basis on hundreds of papers across a wide variety of fields, including databases and computation biology and geometry. If it is indeed false, then we could in theory find better algorithms to solve problems in those fields more efficiently. The scientists who proposed ETH are now

part of the group, opening I asked to examine and explain the results generated by its models. Finally, the physics results involve proof of interacting quantum field theories containing gauge fields and fermions in four dimensions. The group examining the physics results has yet to publish anything about it, but it has provided an email for all questions about its efforts. The company says it will not seek any academic credit for it. models discoveries. Instead, the people in each group will get credit for the write-ups and materials they create. The groups will also be in charge of making sure that the AI-generated results appropriately credit all the researchers, techniques, and ideas the company's models drew on,

end quote. So I kind of don't know what any of that means, but the reaction online from those who do is pretty universal. Here's someone named at underwater Beppis, quote, I've been in shock for the past hour, just stunned. I studied complexity theory a lot in undergrad and spent hundreds, maybe a few thousand of hours on this. The singularity really is happening, isn't it? End quote. Apple this morning unveiled a new 14-inch MacBook Pro with an M6 chip with a 12-core CPU, up to 32 gigabytes of unified memory, and Wi-Fi 7 priced from $1,99.99 bucks available November 6. Also, An updated 24-inch iMac with an M6 chip with up to a 12-core CPU, up to a 12-core GPU,

and a dual 16-core neural engine priced from $1,49, available November 6, quoting 95 Mac. The new M6 MacBook Pro offers Wi-Fi 7 and Bluetooth 6 connectivity up from Wi-Fi 6E and Bluetooth 5.3 on the previous generation model. In terms of ports, you get H-DMI, MagSafe 3, 3 Thunderbolt 4 ports, an SD card slot, and a headphone jack. There's a 12-mapsule center-stage camera. The new MacBook Pro also offers faster charging, reaching up to 50% charge in just 30 minutes when paired with a compatible higher-wattage USBC power adapter, such as the 96-watt power adapter. This is the base model MacBook Pro and Apple's light-up replacing the previous model powered by the M-5 chip. Apple is also still selling the M5 Pro and M5 Max MacBook Pro models, which rumors say will get a huge upgrade as soon as tomorrow with OLED touchscreens.

The new M6 MacBook Pro starts at $1999 with 16 gigabytes of unified memory and one terabyte of storage. It will be available to pre-order tomorrow, October 28th, with the first orders arriving on Friday, November 6th. It's available in silver and space black, end quote. and quoting The Verge. Apple is updating the iMac with M6 chips. The new iMac generation was introduced today ahead of Apple's Pro-Andel event that's scheduled to take place on Wednesday, giving Apple's all-in-one desktop computer a performance boost via the same chip set it launched with the Mac Mini in August. Prices start from $1,49.

And the design otherwise appears largely unchanged. This refresh comes two years after Apple gave the iMac a colorful M4 overhaul, We're skipping the M5 series entirely on this hardware. That should make this upgrade more meaningful. The M6 is Apple's first two-nanometer chip and includes a dual-16 core neural engine that can boost performance when running on-device AI tasks. The M6 configurations also start from a 10-core CPU and a 10-core GPU, sizing up from the 8-core CPU available on the M4 iMac, end quote. So this is interesting because they're sort of bending over backwards to make sure you can't call it layoffs. But Microsoft is apparently offering buyouts to around 60,000 workers, including at a VP level and below across the U.S., Canada, India, and the UK.

So here's the thing. Microsoft's headcount at the end of June was 223,000. So they are essentially attempting to reduce their workforce. by one quarter. That's a pretty big deal, quoting Bloomberg. Divisions such as Xbox, Windows co-pilot, and finance are fully participating in the offer, according to people familiar with the terms. Certain pools of employees, including some focused on data centers, sales, marketing, cybersecurity, and quantum computing aren't eligible, the people said. Microsoft declined to comment on these specifics. This offer gives eligible employees who may be considering a future outside the company a new choice to make that transition now with support. Chief People

officer Amy Coleman said in a note to staff Monday. Workers will receive benefits similar to a standard severance package, the spokesperson said. The world's largest software maker has shed tens of thousands of workers in waves of layoffs in recent years. In July, it offered buyouts to retirement age employees, and more than 30 percent of the roughly 8,750 eligible people took the offer. Microsoft expects far fewer personnel to accept the latest buyout proposal compared with the voluntary retirement program, the spokesperson said. Microsoft's headcount at the end of June was $200,000. $123,000. The figure has been largely flat since 2022, about the time Microsoft began spending heavily on AI and data centers, squeezing budgets, and hiring elsewhere in the company.

The Microsoft spokesperson said the company wants employees who are engaged on the hard problems it's trying to solve. That echoes other tech bosses, who've publicly made clear that continued employment means embracing a faster pace of work and change. Workers deciding whether to take a buyout typically assess their savings and whether they're at risk of layoffs later. Mid-career professionals also consider the job market and how long it might take them to find a new role or whether the package gives them enough runway to make a pivot. Buyouts typically go to more senior, higher cost staff, as with Microsoft's round of early retirement offers, and offer more generous terms than a layoff. In some cases, Microsoft will try to retain top talent if employees opt to take the buyout offer, the spokesperson said. The buyouts give Microsoft units another way to adapt their workforce to changing needs.

Colman wrote in the memo, end quote. Your pricing formula used to be pretty straightforward. More users equals more revenue. But then AI messed with the math. Now each customer gets more expensive to serve as they use your product more. Plus, your product teams are also adding new features, models, and AI agents faster than most billing systems can keep up with. You need a new pricing model, but building that metering and entitlement logic in-house can go from a quick project to a permanent resource-intensive maintenance job faster than you'd expect.

ChargeB can help. They handle the metering, billing, and monetization infrastructure so you don't have to. Get connected with a pricing and monetization expert at chargeby.com slash ride home. That's C-H-A-R-G-E-B-E-E.com slash ride home. AI can clone your voice, forge your signature, and even create fake documents that look real enough to convince an expert. It's called title fraud, and it's how criminals are stealing from homeowners. Here's how it works. Your property records are public by law, which makes it easy for scammers to get your records. Use AI to forge a transfer document and make themselves the new legal owner of your property. Protect your property with home title lock. They monitor your property titles 24-7 and alert you of any changes. If you're a victim of fraud, they'll spend up to $1 million to restore it.

Get a free title history report, plus a 14-day free free. free trial at home titlelock.com slash TBRH. That's home titlelock.com slash TBRH. Finally today, more bits and pieces from the Anthropic S-1. Apparently, as of September 30th, Anthropic had more than 11,900 customers spending more than $100,000 annually with them. They had more than 1,900 customers spending more than 1 million annually and around 250 customers spending more than 10 million annually. Anthropic is apparently aiming to go public the week of November 16th. There were 94 pages of risk factors in the S-1,

and there were warnings that the scale of harm from a catastrophic AI incident might mean it is unable to obtain insurance. Post-IPO, it will keep operating as a public benefit corporation, giving its seven co-founders 50.1% voting control via a single Class F share. And then there's this, quoting Forbes. Contained in the AI Giants IPO filing is an elaborate and somewhat unusual succession plan, confidential lists of people who could take over its founder's powerful voting roles. According to a filed Anthropic Founders LLC agreement attached to the company's S-1,

each founder can nominate a ranked list of up to 10 successors subject to approval from five of the other six co-founders who will take over their votes. Succession provisions are common at founder-controlled companies, for instance, Facebook's IPO specified a successor designated by Mark Zuckerberg, end quote. We haven't read our old friend Matt Levine in a while, but I enjoyed his take on all of this. Quote, over the weekend, Bloomberg published an essay by Tom Orlik, Michael Deng, and Dong Mayo, Zhang, titled Anthropics Pitch to Investers, starts with taking their jobs. Buy our IPO, we will use the funds to take your job and maybe kill you, doesn't immediately sound like an attractive offer. In a strange way, though, it has a perverse appeal.

Smart investors understand the importance of hedging against risk. If you can't beat our new AI overlords, why not join them? Maybe use the profits to build a bunker. That's right, right? The biggest risks in Anthropics' prospectus are not really risks to its business. They are risks that its business poses to everyone else. The one I quoted above shugs into its conclusion that killing everyone on Earth, quote, could adversely affect our business, financial condition, and results of operations.

But surely that's not the main problem with killing everyone on Earth. Or another risk factor says, quote, as AI systems become more capable or prevalent public perception that our models, platform, and applications contribute to job displacement or economic disruption could generate significant negative media coverage, customer or employee backlash or organized opposition regardless of the actual impact of our models, platform, and applications on employment. The explicit content of that one is,

we might get negative media coverage and customer backlash, but its implicit warning is AI might take everyone's jobs. In the maximalist model of AI adoption, the one in Anthropics risk factors, all other human enterprise will be eaten by Claude, so you might as well own some Claude. The more dangerous Anthropic is to everyone, the more important it is to be on Anthropic side. An Anthropic has a complicated governance structure

with five classes of common stock. We talked last week about the traditional method of maintaining founder control of a tech company. It's just the founders get special stock with 10 votes per share, The math on that is that as long as the founders own at least 9.1% of the company, they will have a majority of the voting power and will be insulated from outside shareholders. There's no magic to these numbers. They're just a rough way to balance giving the founders an extra say over the company they created

with making sure they have a reasonable amount of skin in the game. But if you're really concerned about AI risk, that is probably insufficient in two ways. On the one hand, if Anthropic needs to raise a gazillion dollars to safely build AI, there's no guarantee that its founders will continue to hold at least 9.1% of its economic value. Divorcing governance from economic value entirely giving the founders majority control, no matter how much stock anthropic cells is a way to ensure that the trustworthy, public-spirited founders will always keep control of their robot army. On the other hand, how much do you trust those founders?

The robot army line in the previous paragraph was a reference to Elon Musk, who has said, I just don't feel comfortable building a robot army at Tesla without significant voting control, and who will probably consolidate that voting control by merging Tesla with SpaceX. Elon Musk surely trusts himself to command a robot army, and many Tesla and SpaceX shareholders probably agree, but not every outside observer might, giving a nonprofit board made up of non-billionaires with no economic ties to the company. The power to supervise the founder's command of their robot army might help a little,

though, of course, there's no guarantee that the nonprofit trustees will do the right thing either. Maybe they and the long-tenured employees, will team up to seize control of the robot army. The trustees have no economic interest in anthropic stock, but money isn't everything, and it may be difficult to know what role money will play in a post-AGI world. It is possible that I am overthinking this. This is the governance structure that you would put in place if you worried about existential AI risk, but it is also the governance structure that you would put in place if you ran a fun, quirky little mission-driven worker cooperative that happened to need to raise hundreds of billions of dollars of outside capital. Sure, we'll

sell stock, you'd say, but we don't want to sell stock. We want to keep our friends in charge and not do anything that would make our jobs less fun or more business-like. This is the way to run a worker's co-op at a public company, end quote. And finally, I've seen several takes like this that the filing is sort of a humble pie moment for AI skeptics, quoting the information. One item stood out to me, how Anthropic is running its models more efficiently than before. Anthropics' ability to generate $3.5 billion in free cash flow in the first half of 2026 from $16.3 billion in revenue shows just how much it has improved the way it runs AI in the cloud compared to the same period a year earlier when it burned $1.6 billion. Specifically, Anthropics said it is generating 85% of gross profit margin from powering existing models, which doesn't include the cost of developing new models. Anthropic is doing so even as it relies almost entirely on computing it rents from cloud providers,

which themselves need to generate a profit. So its IPO filing is a reality check on the AI skeptics' longtime view that a business model founded on selling models on a pay-as-you-go consumption basis wouldn't be profitable, end quote. Nothing more for you today. Talk to you tomorrow.