← Tech Brew Ride Home
Tech Brew Ride Home
Ouroboros All The Way Down?
00:19:43
Nvidia is investing $100B in OpenAI, and there are a billion angles to that fact, so get ready to dive into it. Why were there giant sim farms popping up around NYC? The new AI to help you with Candy Crush, I guess. And new buzzword alert: say hello to “workslop.”
Loading summary
Transcript5 lines
- [00:04]
A
Welcome to the Tech Brew Ride home for Tuesday, September 23rd, 2025. I'm Brian McCullough. Today, Nvidia is investing $100 billion in OpenAI, and there are at least a billion angles to that fact. So get ready to dive into it. Why were there giant SIM farms popping up around New York City? The new AI to help you with Candy Crush, I guess, and New Buzzword Alert. Say hello to Workslop. Here's what you missed today in the world of tech. Nvidia is going to invest $100 billion in OpenAI. It's going to do so progressively as part of a deal to deploy at least 10 gigawatts of Nvidia systems for OpenAI's infrastructure in the coming years. Quoting the Verge One of the most important points here, besides more data centers and compute, which are always in high demand for companies like OpenAI, is that as part of the deal, Nvidia quote intends to invest up to $100 billion in OpenAI progressively as each gigawatt is DEP. Per the release. The details will be finalized in the next few weeks. According to the companies, Nvidia will now be a preferred strategic compute and networking partner for OpenAI's AI factory growth plans. OpenAI said news of this strategic partnership comes after the January announcement that Microsoft would no longer be the sole compute provider for OpenAI and that the tech giant would now just have a right of first refusal in such cases. Since then, OpenAI has branched out, beginning to build its own data Centers and a $300 billion cloud computing deal with Oracle. OpenAI and Microso also put out a pointed statement earlier this month that they had signed a non binding memorandum of understanding, adding that they are actively working to finalize contractual terms in a definitive agreement. End quote and quoting CNBC Sam Altman and Jensen Huang negotiated their pact largely through a mix of virtual discussions and one on one meetings in London, San Francisco and Washington, D.C. with no bankers involved, according to people close to the talks who declined to be named because they weren't authorized to speak publicly on the matter. The arrangement calls for Nvidia to invest 10 billion doll at a time in OpenAI, the company behind ChatGPT. As the buildout unfolds, Nvidia will also supply the cutting edge processors powering a host of new data centers. Now, while OpenAI gets more intimate with Nvidia, it has to maneuver through a number of high stakes relationships with other key partners. OpenAI only informed Microsoft, its principal shareholder and primary cloud provider, a day before the deal was signed, the people familiar with the matter said. Earlier this year, Microsoft lost its status as OpenAI's exclusive provider of computing capacity. The pact also comes less than two weeks after a disclosure from Oracle indicated that OpenAI agreed to spend $300 billion in computing power with the company over about five years starting in 2027. At the start of the year, OpenAI joins Stargate, a multibillion dollar project announced by President Trump and backed by Oracle and SoftBank, to build out next generation AI infrastructure. Going forward, all of OpenAI's infrastructure projects will fall under the Stargate umbrella. Nvidia and OpenAI provided scant details about where and when the buildout will take place, other than to say that the first 10 gigawatt go online in the back half of next year. Executives said They've reviewed between 700 and 800 potential locations since unveiling Stargate in January. In the months that followed, they fielded a flood of proposals from developers across North America offering land, power and facilities. That list has been narrowed as OpenAI weighs energy availability permitting timelines and financing terms, the company said in Monday's announcement. OpenAI described Nvidia as a preferred partner, but executives told CNBC that it's not an exclusive relationship and the company is continuing to work with large cloud companies and other chip makers to avoid being locked into a single vendor. The financing structure for the OpenAI deal is designed to avoid hefty dilution. The initial $10 billion tranche is locked in at a $500 billion valuation and expected to close within a month or so once the transaction has been finalized, people familiar with the matter said. Nine successive $10 billion rounds are planned, each to be priced at the company's then current valuation as new capacity comes online, they said this morning. Sam Altman said OpenAI wants to create, quote, a factory that can produce a gigawatt of new AI infrastructure every week and plans to reveal more details this year. Jensen huang said the 10 gigawatt OpenAI project is equivalent to 4 to 5 million GPUs. The first phase is expected to come online in the second half of 2026 using Nvidia's Vera Rubin platform, and sources say the Nvidia OpenAI deal is really two separate transactions. Nvidia invests in OpenAI for non voting share, then OpenAI can use the cash to buy Nvidia's chips and that is what a lot of people are talking about already. The whole AI mania has been accused of being a bubble, in part because all of the investment money is pouring into AI, often from the big tech players who then receive the money back in terms of AI compute spend in a snake eating its tail sort of way. Well, I will point out this is a $100 billion investment on Nvidia's part. I know it's over time but but let me point out, do you know how much money Nvidia has made in net income over the last three years? $107 billion. Are they just turning around and spending all the money they made on one of their biggest customers? Or vice versa? Or maybe similarly OpenAI is burning a ton of money. We know it needs more money if it's going to keep burning it and it burns it by spending it on data centers on Nvidia chips. So one way to look at this is Jensen is giving Sam the money to pay for the chips Sam needs to buy from him. Another thing to point out though, this at least at the contours described thus far, represents as much COMPUTE as Nvidia shipped this year in total. Would be useful for them to lock in that demand right from potentially their biggest long term customer. So is this Ouroboros all the way down now there's a metric ton of takes out there on this right now, so let me give you a sampling just to get a sense of the discourse. Connor Sen tweeted, I'm not sure if I'd classify this as a bubble, but 10 companies passing tens or hundreds of billions of dollars back and forth is certainly something. Here's Reihard Jark for Both Nvidia and OpenAI, the $100 billion Nvidia investment is perfect for OpenAI. The big question was how they were going to raise the future $300 billion as the valuation is already very high and the cash burn over the next few years is projected to be crazy. On top of it, there is not that much VC money that size. Nvidia, because of their other motivations in this deal, will also not be very demanding when it comes to the valuation I would bet. But on Nvidia's part, they want to make sure OpenAI continues to buy a ton of Nvidia GPUs as they don't want the AI train to stop. OpenAI and Anthropic are already now 1/3 of Nvidia's GPU purchases. The biggest risk for Nvidia was that the capital and venture markets saying stop to all that spending. So to remove the pressure from capital markets and VC money, they decided to just secure their future revenue with their own cash that they don't know what to do with for Nvidia but it is also best that OpenAI wins in this AI race as Google with its TPUs is not a perfect customer and Anthropic because of its Amazon backing is also using and testing out Amazon Trainium chips. The biggest OAI backers Microsoft and Oracle don't have or have very early ineffective custom ASIC projects so they are not a threat. Nvidia with a considerable stake in OpenAI can also influence OpenAI's future decisions about their own custom silicon End quote Here's Tuchin Jin on X Sam altman I need GPUs Jensen Huang you know who has GPUs Sam but I don't have any money. Jensen let's do a strategic partnership 9am Nvidia to invest $100 billion in OpenAI to deploy 10 gigawatts of data centers 9:20am Nvidia's stock rockets by more than $100 billion Jensen here's my $100 billion investment and Shea Balor on X the bears say Nvidia is just buying 100 billion DOL own chips through OpenAI. The reality is that $100 billion is 10 gigawatts of capacity. The power of New York City locked into Nvidia stack. That's Blackwell today, Rubin tomorrow, and years of high margin networking plus software on top, end quote the US Secret Service says it has dismantled more than 300 SIM card servers in the New York City area that could have disrupted communications ahead of the upcoming UN General Assembly. Quoting Bloomberg agents discovered more than 300 SIM card servers and 100,000 SIM cards at several locations within a 35 mile radius of new York City, according to a statement on Tuesday. The Secret Service didn't identify the officials who were threatened, the nature of the threat, or the nations that may have been involved, they said the Secret Service hasn't publicly called out any nation in part because of sensitivities around this week's UN General assembly meeting. The potential for disruption to our country's telecommunications posed by this network of devices cannot be overstated, secret Service Director Sean Curran said in the statement. He said the investigation, quote, makes it clear to potential bad actors that imminent threats to our protectees will be immediately investigated, tracked down and dismantled, end quote. Bad actors using SIM card servers and thousands of cards can send anonymous threats and conduct attacks such as disabling cell towers and overwhelming communications networks. Matt McCool, the special agent in charge of the Secret Services New York field office, said the seizure was the largest operation of its kind the agency had undertaken. Agents were still conducting forensic analysis on the more than 100,000 SIM cards, he said. The timing, the location, the proximity of this network had the potential to impact the UN and that was clear and something that we had to consider, McCool said. Initial analysis of the seized data has identified ties to at least one foreign nation, as well as links to criminals already known to US Law enforcement, including cartel members, according to people familiar with the investigation. It wasn't clear what connection the foreign nation and the criminals had to each other, if any, or what their links were to the devices. It was also unclear if the network was linked to a spate of incidents earlier this year that saw unknown actors impersonate White House Chief of Staff Susie Wiles and Secretary of State Marco Rubio over the summer, a State Department cable said an unknown person left voice and text messages for at least five people, including three foreign ministers, a US Governor and a US Member of Congress, after creating a signal account that pretended to be Rubio's in mid June. Britain has already taken steps to restrict SIM farms. In April, the home office announced a ban on the possession or supply of SIM farms without a legitimate reason, citing their role in smishing campaigns that impersonated delivery services or banks. End quote.
- [11:01]
B
I'm Christian McCaffrey, pro running back, and Abercrombie is an official fashion partner of the NFL. I'm not kidding when I say NFL by Abercrombie broke the Internet last year and I think this season's lineup is even cooler and so does my wife who keeps stealing all my hoodies. Stay fit for the season and Abercrombie's newest arrivals shop NFL by Abercrombie in the app, online and in store.
- [11:31]
A
Curious about the digital infrastructure that powers today's biggest tech? Then check out the Interconnected podcast. It's a new series from Equinix that explores emerging tech megatrends and the infrastructure, making them possible. Through candid conversations with industry experts, the show breaks down highly complex topics in a way that's easy to understand. In the very first episode covering AI in the medical field, the hosts do a great job helping listeners understand the computational models and interconnected data systems powering today's most promising medical developments. From there, they reveal the connection between this infrastructure and the groundbreaking healthcare breakthroughs we're seeing lately, like rapid diagnostics, personalized medicine, and supercharged drug discovery. Stay tuned into the digital infrastructure powering today's biggest tech trends with Interconnected Give it a listen and follow the show on YouTube, Apple, Spotify or wherever you get your podcasts.
- [12:27]
C
When did making plans get this complicated? It's time to streamline with WhatsApp, the secure messaging app that brings the whole group together. Use polls to settle dinner plans, send event invites and pin messages so no one forgets mom 60th and never miss a meme or milestone. All protected with end to end encryption. It's time for WhatsApp message privately with everyone.
- [12:51]
A
Learn more@WhatsApp.com Google this morning announced Play Games Sidekick for Android, an in game overlay that provides access to Gemini Live while playing games downloaded from the Play Store. Quoting 9to5Google Google is bringing Gemini Live to Android Gaming with Play Games Sidekick build as a coach. It's meant to be a helpful in game overlay available while playing games downloaded from Play. In supported games, you'll see a new handle that can be dragged to either side of the screen or dismissed and accessed from notifications. This overlays a panel which supports both landscape and portrait orientation with a link to your profile and play points. Next is a grid to take a screenshot screen recording stream to YouTube and enable. Do not disturb. Then there's a prominent Start Gemini Live button that launches the assistant and starts sharing your screen. Inline controls let you mute the microphone and end. You can ask anything about the game with Gemini fully aware of the context. The idea is to get help without having to leave the game to do a web search for walkthroughs or other tutorials you get in game guidance with verbal tips, tricks and support. The context was perfect when Google demoed Sidekick with the Battle of Polytopia. For us, Google Play worked with the Google DeepMind team with the last first demoing AI agents for games last December. Gemini can reason about the game based solely on the action on the screen. Meanwhile, game tips in Sidekick is powered by AI and based on your gameplay. There are also carousels for achievements which you can browse, streaks, quests, coupons, play credit which is redeemable in line and boosts. Google will start letting users try Play Games Sidekick with Gemini Live in select games over the coming months. Finally today, more intriguing data around the use of AI. According to new research, those low productivity gains statistics from AI that many companies have been reporting thus far may stem from employees using AI to produce work slop or low effort passable work that creates more work for others. So new buzzword for you workslo quoting Harvard Business Review A confusing contradiction is unfolding in companies embracing generative AI tools. While workers are largely following mandates to embrace the technology, few are seeing it create real value. Consider, for instance, that the number of companies with fully AI led processes nearly doubled last year, while AI use has likewise doubled at work since 2023. Yet a recent report from the MIT Media Lab found that 95% of organizations see no measurable return on their investment in these technologies. So much activity, so much enthusias, so little return. Why? In collaboration with Stanford Social Media Lab, our research team at Better Up Labs has identified one possible reason employees are using AI tools to create low effort, passable looking work that ends up creating more work for their coworkers. On social media, which is increasingly clogged with low quality AI generated posts, this content is often referred to as AI slop. In the context of work, we refer to this phenomenon as work slop. We define work slop as AI generated work content that masquerades as good work but lacks the substance to meaningfully advance AI given task. Here's how this happens. As AI tools become more accessible, workers are increasingly able to quickly produce polished output, well formatted slides, long structured reports, seemingly articulate summaries of academic papers by non experts, and usable code. But while some employees are using this ability to polish good work, others use it to create content that is actually unhelpful, incomplete or missing crucial context about the project at hand. The insidious effect of workslop is that it shifts the burden of the work downstream, requiring the receiver to interpret, correct or redo the work. In other words, it transfers the effort from creator to receiver. If you have ever experienced this, you might recall the feeling of confusion after opening such a document, followed by frustration. Wait, what is this exactly before you begin to wonder if the sender simply used AI to generate large blocks of text instead of thinking it through. If this sounds familiar, you have been work slopped. According to our recent ongoing survey, this is a significant problem. Of the 1150 US based full time employees across industries, 40% of report having received workslop in the last month. Employees who have encountered workslop estimate that an average of 15.4% of the content they receive at work qualifies. The phenomenon occurs mostly between peers. 40% but workslop is also sent to managers by direct reports at 18%. 16% of the time workslop flows down the ladder from managers to their teams or even from higher up than that. Workslop occurs across industries, but we found that professional services and technology are disproportionately impacted. End quote hey, quick reminder that this pod has a YouTube channel. Just search tech brew ride home in YouTube and subscribe. The reason you're going to want to do that is because all of the bonus interview episodes going forward will also have video as well, so you're gonna want to subscribe to see that. For example, this weekend's upcoming bonus episode was actually recorded in person in the Morning Brew Studios. And while we're on the topic, reminder that you can subscribe to the ad free podcast feed at Tech Supercast Tech. Talk to you tomorrow.