Oct 28, 2026 · 8 min listen · Last updated October 28, 2026
From storyflo. This is your daily audio brief. Theo here. October 28th, tech desk. Five stories from the last twenty-four hours — here's where I'd start. Let's get into it. First, from mtslive. 10/27: AI Disproves the Exponential Time Hypothesis.
Listen · storyflo · tech
Daily Tech Brief · October 28th
0:00-8:02
Pick your daily storyteller
Subscribe to match with Theo, Jessica, Chloe, Mason, Brock — your voice, every brief.
Audio pre-rendered by Storyflo · cached + delivered from the edge
10/27: AI Disproves the Exponential Time Hypothesis
Hey there! So, there’s been some fascinating movement in AI lately. OpenAI's models have made waves by disproving the Exponential Time Hypothesis, which is a big deal because it challenges a long-held assumption about how tough certain computational problems are. This could shake up a lot of foundational work in computer science. Meanwhile, Anthropic's Claude model also claims to have found a new algorithm that can tackle some of these hard problems in less time, which, if validated, could really change the landscape of computational theory.
On the math front, Axiom's AI agents helped verify an extensive proof in record time, converting a massive 15,000 pages into a manageable 43 million lines of code. They even found new insights while fixing gaps in existing proofs. It’s impressive how AI is collaborating with humans to push boundaries in mathematics.
In biotech, Anthropic has acquired Mithril Medicine, a startup that’s been using AI to develop treatments for rare diseases. The founder, Cody Tranbarger, has been leveraging AI to run multiple drug programs simultaneously, which is pretty innovative.
And then there’s Google’s Genie 4, which can create interactive environments in real-time, allowing users to manipulate the world as they explore it. This could open up so many possibilities for simulations and training AI agents.
Lastly, Oracle is expanding its AI cloud offerings with new chips and is investing heavily in infrastructure, while a robotics startup is raising significant funds despite not having revenue yet. It’s all a bit wild, isn’t it?
This might just blow your mind. I found a GitHub repo that can become your AI job assistant. And there is proof it works. It’s called Career Ops (cool, memorable name in my opinion). The creator, Santiago Fernández de Valderrama Aparicio, says he evaluated 740 job listings, applied to 68, got 12 interviews, and landed one offer. Those are his reported results, not a promise about yours. His repo and story. I like the idea behind it: spend less time rewriting your resume for jobs that were never a good fit. A smaller shortlist. Better evidence. Your decision.
‘Racing Far & Wide’ in AI. Anthropic, Nvidia & Google. ARD #193
Today’s AI landscape is buzzing with activity, especially with Anthropic, Nvidia, and Google making significant moves. Anthropic is showing impressive efficiency in AI operations, highlighted by its recent S-1 filing. They’ve managed to cut their compute costs dramatically, thanks to innovative algorithms and a diverse chip strategy, which allows them to maximize performance across different hardware. This efficiency has led to a remarkable gross margin of around 85%, a stark contrast to their previous losses.
Meanwhile, Nvidia is investing $3 billion in Meta’s stock sale, strengthening their partnership to enhance AI agents. This collaboration will utilize Nvidia's cutting-edge infrastructure, crucial for developing Meta's Muse personal AI agent app. It’s interesting to see how Nvidia positions itself as a central figure among major tech players, tailoring its capabilities to fit various needs.
On another front, Google is diving into world models with its new Genie 4 model, which goes beyond traditional large language models. This model simulates a 3D environment with realistic physics and interactive characters, offering a more immersive experience. Google’s ambition in this area is notable, especially as it aims to leverage its vast distribution channels across industries and consumers alike.
Overall, the race in AI is not just about speed but also about efficiency and innovation. As companies like Anthropic and Google push the boundaries, it’ll be fascinating to see how this dynamic unfolds in the coming months.
AI: Anthropic’s S-1, more on “Mo’ Money & Mo’ Problems”. AI-RTZ #1253
It’s here. Anthropic filed its S-1 with the SEC on Monday, for its long-expected mega-AI IPO on the Nasdaq, under the ticker ‘ANT’. It is the second ‘mo’ money’ mega-AI IPO of this AI Tech Wave, after Elon Musk’s SpaceX in June, and ahead of arch-rival OpenAI, now expected next year. Anthropic, once the Pepsi to OpenAI’s Coke, has turned into the Coke, for now. Back in June, when the draft was filed confidentially, I noted in AI-RTZ #1105 that all three mega-AI IPO candidates had filed, with my former firm Goldman Sachs on the lead left of each one.
AI slows growth for Infosys and keeps it afloat at the same time
Indian tech services giant Infosys has mixed news about AI. In its most recent quarterly results late last week, the company revealed that it’s racked up 29 percent sequential growth in AI revenue, which now delivers 10.5 percent of Infosys’ incoming cash. Those are far nicer numbers than the 1.5 percent year-on-year revenue growth Infosys reported for its second quarter. “Growth was below our expectations. Volume was softer, and price increases were lower than anticipated amid high competitive intensity,” chief financial officer Jayesh Sanghrajka told investors.
A researcher claims Samsung TVs use connected devices with mics to listen in on you
It seems people have been tinkering with their smart TVs a lot lately. Over the past few days, we've seen people using Claude Code to remove ads and add new home screens. Now, someone has gone a little bit further and claims they've discovered proof that Samsung TVs that don't have a microphone will instead try to connect to one that does to listen to you.
Cisco’s partner shift: What customers should expect from their Cisco partner in the AI era
Most of the coverage from Cisco Partner Summit will understandably focus on what Cisco is doing for its more than half a million partners. There were new incentives, a revamped deal registration program, and plenty of talk about margins. But I’ve always believed the channel story is really a customer story. When I was a Cisco customer, I relied heavily on my partner for guidance and support. Roughly 90% of Cisco’s bookings flow through partners, so for most network and security leaders, the partner is the Cisco experience.
Airborne micro particles can be a significant health hazard in a workshop, and thus it’s important to maintain a high standard of ventilation. But monitoring their concentration is also important, and thus there exists a range of sensors to help keep you safe. [WeSpeakEnglish] has taken this idea to the MAX, with a PM2.5 sensor that drives a very large wall mounted display.
Etzioni on AI: Is Anthropic worth $2 trillion? Do the napkin math with this handy calculator
Anthropic filed to go public on Monday under the ticker ANT, and you may end up owning the stock soon even if you never buy it directly, because index funds and ETFs will be required to buy it within weeks of its debut. Anthropic’s bankers are reportedly valuing Anthropic between $1.5 trillion and $2 trillion. Below is a simple, interactive tool to help you form your own opinion of what’s an appropriate valuation for Anthropic. Disclaimer: this is highly simplified, and I’m not an investment advisor.
VCs will make billions on Anthropic’s IPO, but Amazon is set to make much (much) more
So, here’s the scoop on Anthropic’s upcoming IPO. It’s shaping up to be quite the payday for venture capital firms like Menlo Ventures and Sequoia, but they’re not the biggest winners here. Surprisingly, none of these VCs own a significant enough stake to be listed as principal stockholders. That’s unusual for a tech IPO, especially given Anthropic’s massive valuation of $965 billion. For context, even Menlo's $14 billion stake is just about 1.5%.
The real financial powerhouse in this scenario is Amazon. They first backed Anthropic in 2023 with a commitment of up to $4 billion and have since invested heavily, structuring much of it as convertible notes. As of June 30, they had $6.2 billion in principal outstanding, which is Anthropic’s entire debt load. But here’s the kicker: those notes are now valued at nearly $98 billion. So, when they convert, Amazon could potentially snag a huge chunk of shares, although there’s a cap limiting them to 12% of the total share count.
On top of that, Amazon has also poured $10 billion into nonvoting preferred stock, bringing their total investment to around $18 billion. They’re not just looking at stock gains, either. Anthropic has committed to spending at least $110 billion on AWS over the next decade, which means Amazon will continue to rake in revenue from them.
Google’s in the mix too, with a stake that’s been diluted but still significant, and they’ve secured cloud commitments worth $114 billion through 2033. It’s fascinating to see how these tech giants are positioning themselves, especially since they’re also competitors in the AI space. The S-1 hints at potential conflicts of interest, but for now, it looks like Amazon and Google are set to benefit the most from Anthropic’s IPO.