← Tech Brew Ride Home
Tech Brew Ride Home
OpenAI Has Gone For-Profit
00:23:37
OpenAI has officially done its whole reorg to for-profit thing. Amazon is cutting a metric ton of employees and its blaming AI. Elon launches Grokpedia to take on Wikipedia. And does Anthropic’s success in the enterprise give it a leg up on OpenAI in the one metric that counts? Making money?
Loading summary
Transcript6 lines
- [00:04]
A
Welcome to the Tech Brew Ride home for Tuesday, October 28, 2025. I'm Brian McCullough. Today OpenAI has officially done its whole reorg to for profit thing. Amazon is cutting a metric ton of employees and is blaming AI. Elon launches Grokopedia to take on Wikipedia and does Anthropic's success in the enterprise give it a leg up on OpenAI in the one metric that counts, which is making money. Here's what you missed today in the world of tech. If you're looking for enterprise grade identity automation minus the enterprise grade baggage, aka having your users log in 500 times, Yeshid delivers advanced IAM automation without moving teams onto a legacy identity provider. Whether you use Google Workspace, Microsoft 365 or Okta, Yeshid integrates directly. No rebuilds or RIP and replaces are required. Yeshid helps IT and security teams reduce risk, not just tickets, and IT teams everywhere might have just breathed a collective sigh of relief. Every access, change, review and approval is tracked and exportable, helping security teams effortlessly demonstrate compliance with SOC2, ISO or HIPAA. IT and security teams can spot risk before it becomes a finding. Learn more@yeshid.com techbrew that's y dashes h I d.com techbrew OpenAI says it has, quote completed our recapitalization the OpenAI foundation now has equity in OpenAI valued at around $130 billion and it continues to control the OpenAI for profit entity. Microsoft says it supports this move, got a 27% stake in the form of a $135 billion Public Benefit Corporation investment. Also got access to OpenAI's AI models until 2032, including AGI level models, and will buy $250 billion worth of Azure services. Quoting the Verge, the company's for profit arm is now a public benefit corporation dubbed OpenAI Group PBC. The nonprofit is now called the OpenAI foundation and quote holds equity in the for profit currently valued at approximately $130 billion. It will begin with a $25 billion focus on healthcare and disease and AI resilience. Per OpenAI's blog post, the nonprofit will also get additional ownership after OpenAI's for profit reaches an UNSP valuation milestone. The news comes after more than a year of OpenAI's negotiations with the offices of the Attorneys General of California and Delaware. If they hadn't eventually blessed the restructuring, OpenAI wouldn't have been able to move forward. It also follows a thorny and drawn out legal battle with Elon Musk, who has been suing both the company and CEO Sam Altman in attempts to stop the conversion. Musk Co founded OpenAI in 2015 as a nonprofit research lab. One question that sparked controversy over the past year and still remains not fully answered is whether OpenAI's nonprofit entity will still retain control over its underlying technology, including the potential future development of artificial general Intelligence, or AGI, systems that equal or surpass human cognitive ability. AGI is the moving goalpost that OpenAI and nearly all of its competitors have been chasing, funneling increasing financial resources and headcount into its development. If OpenAI didn't announce the completed restructuring by New Year's Eve, it could have lost up to $10 billion of its previously announced softbank investment. And quoting Bloomberg, the foundation plans to initially focus on funding work to accelerate health breakthroughs, among other efforts. Microsoft shares jumped almost 4% in early trading. Many on Wall street have cited the changing OpenAI relationship as a serious point of uncertainty for the software maker. One major sticking point in the months long negotiations between Microsoft and OpenAI had been what happened once OpenAI had achieved AGI or AI that outperforms humans at most economically useful tasks. Under the new agreement, the threshold must be verified by an independent expert panel. Once achieved, Microsoft will no longer get a cut of OpenAI's revenue. Microsoft is also losing its right of first refusal on new cloud infrastructure business from OpenAI. Azure had long been OpenAI's exclusive provider, but Microsoft began allowing it to seek out services from other vendors like Oracle, as long as it had an option to the business first. OpenAI will make an additional $250 billion commitment to Azure. Microsoft's access to OpenAI's technology will not include consumer hardware, the company said. OpenAI will also have the ability to jointly develop some products with third parties. Amazon says it will cut around 14,000 jobs from its corporate workforce after cutting 27,000 across 2022 and 2023. The company had 1.55 million employees globally in June. Quoting Bloomberg, this announcement comes just months after Chief Executive Officer Andy Jassy warned that AI will shrink the company's workforce. The downsizing marks Amazon's second round of reductions in about as many years. Jassy signaled in June that the company's staff count would likely fall as it increases its use of artificial intelligence to complete tasks normally handled by people. Those comments touched off panic among workers who trolled anonymous online chat rooms for insights about potential job cuts. The cuts at Amazon will span roles from logistics and payments to video games and the cloud computing unit, according to people familiar with the matter. The reductions we're sharing today are a continuation of this work to get even stronger by further reducing bureaucracy, removing layers and shifting resources to ensure we're investing in our biggest bets, beth Galetti, senior vice president of people, experience and technology at Amazon, said Tuesday in a blog post. The company will increase hiring in some parts of the business, and the 14,000 number was an overall workforce reduction, Galletti said. Reuters was reporting earlier that as many as 30,000 people would lose their jobs. Cuts on that scale would surpass the rolling reductions in late 2022 and early 2023 that ultimately ensnared more than 27,000 corporate employees as Jassy looked to reduce costs after a pandemic era boom. Since then, there has been a steady drip of more modest layoffs targeting individual teams. While Amazon employed a total of about 1.55 million people as of June 30, most of them work in warehouses. The corporate workforce comprises about 350,000 personnel, meaning the 14,000 cuts announced Tuesday represent about 4% of that headcount. Microsoft Chief Commercial Officer Judson Althoff told employees that AI tools were boosting productivity in everything from sales and customer service to software engineering, Bloomberg reported in. Altoff said that AI saved the company more than $500 million last year in its call centers alone and increased both employee and customer satisfaction, according to a person familiar with the matter. End quote Grokopedia, a Grok based encyclopedia, has launched with more than 885,000 articles drawing from Wikipedia content. Elon Musk called it a massive improvement over Wikipedia, quoting the post. The site resembles Wikipedia in style and format with articles on topics such as ChatGPT, Diane Keaton and the 2026 FIFA World Cup. But it appears significantly smaller, more opaque in its workings and more right leaning in how it framed some articles. Grokopedia's entry on gender, for instance, begins with the sentence gender refers to the binary classification of humans as male or female based on biological, sexual. Wikipedia starts with Gender is the range of social, psychological, cultural and behavioral aspects of being a man or boy, woman or girl, or third gender. The project is Musk's latest bid to harness Grok, the ChatGPT like AI system developed by his company Xai to offer right leaning, freewheeling alternatives to popular mainstream tech tools. Some Musk admirers greeted its debut with excitement Monday, while critics highlighted examples of articles that contained falsehoods or passages that copied Wikipedia verbatim. Musk's own Grokipedia entry differed strikingly from Wikipedia's page about him. It describes some of his pursuits in breathless terms, saying his pushes for artificial intelligence emphasize AI safety through truth oriented development rather than heavy regulation, and that certain releases reflect xai's rapid iteration, with Musk highlighting Grok's design for maximal truth seeking and reduced censorship. Grokopedia cited the website of xai, which Musk owns, to make that point. On the section about Musk's work in the US Doge service, Grokopedia included an error regarding Ohio gubernatorial candidate Vivek Ramaswamy, asserting that he assumed a more prominent role after Musk's departure in May. In fact, Ramaswamy left the group before it became part of Trump's administration in January. To support the false claim, Grokopedia cited articles from the BBC and Al Jazeera, neither of which mentioned Ramaswamy. In contrast to Wikipedia's Accolades section, Musk's Grokipedia page concluded with a section titled Recognition and Long Term Vision that expounded on his beliefs. Quote his Long Term Vision prioritizes safeguarding human consciousness against existential threats, emphasizing the establishment of a self sustaining, multi planetary civilization as a hedge against planetary scale catastrophes on Earth, it said. At launch, Grokipedia's homepage boasted that the site had about 885,000 articles, whereas the English language Wikipedia has more than 8 million. The site went live on Monday without fanfare or explanation, a week after the date Musk had initially set for its launch on October 20. He said he was postponing the launch to the end of that week, explaining we need to do more work to purge out the propaganda. A minimalist homepage bore the title Grokopedia version 0.1 and a search bar where users could type in queries. About an hour after it went live on Monday, the site went down, resurfacing later Monday evening. Grokipedia's articles appeared to be derived from the same large language model that underlies the Grok chatbot on X, formerly Twitter, which Musk purchased and renamed in 2022. That could mean it has access, at least in theory, to the latest X posts from the site's hundreds of millions of users, which it can use to inform articles and keep them up to date, end quote. The digital world is more connected behind the scenes than you may realize. Interconnected is a video podcast series by Equinix that explores the hidden infrastructure behind our connected future. From data centers to cloud ecosystems to the platforms and people who use them, Interconnected's hosts bring tech leaders, industry experts and innovators together in candid conversations to break down and discuss the future of global connectivity. The third episode of Interconnected, for example, covers the digital infrastructure for a food secure world. They discuss how farmers are moving from 20th century operations to AI and machine learning that analyzes soil, weather and crop data to tackle 21st century risks, plus how digital platforms are now connecting local producers to global demand through cloud. Linked supply chains follow Interconnected on Apple, Spotify or wherever you get your podcasts. When your data goes dark, Veeam turns the lights back on Partner with Veeam to increase your data resilience and get the right data recovery options for any kind of disruption so you can undo the unpredictable and get your data back so fast you won't even have time to miss it. With Veeam, it's all good. Keep your business running. @veeam.com Adobe held its Adobe Max event where they unveiled AI tools for Photoshop, Premiere Pro and Lightroom and said Firefly Image 5 can now generate images in native 4 megapixel resolution. Adobe is partnering with YouTube to offer a create for YouTube shorts feature in the Premier mobile app, letting creators instantly publish video. Launching soon Adobe plans to support custom AI models in Firefly, updated Firefly to generate soundtracks and speech, and added an AI assistant to Photoshop and Express. They also unveiled Project moonlight, an AI agent on its Firefly platform designed to act as a creative director for social media campaigns via text prompts. Quoting the Verge to start, Adobe is allowing Photoshop users to power generative fill capabilities using Google and Black Forest Lab's third party AI models. After selecting their image and giving generative fill a prompt, such as describing an object to insert, or replacing an existing object or person with something else. Users can switch between Google's Gemini 2.5 Flash Black Force Flux 1 context and Adobe's Firefly image model, providing a wider variety of results to choose from. Photoshop for Web is also introducing an AI assistant in private beta. This is a chatbot like interface that can be given descriptive instructions like Increase the saturation to automatically edit files effectively using Photoshop's vast selection of tools for you. This capability was teased in April and a similar feature is launching for Adobe Express in public beta. Another beta feature called assisted Culling is launching for Adobe Lightroom that can filter large collections of photographs for different focus angles and degrees of sharpness, and recommends the best snaps for photographers to edit. The Firefly AI image model that powers these editing tools is getting a model upgrade. Adobe says that Firefly Image 5 can generate images in a native 4 megapixel resolution without upscaling and has been optimized to improve its ability to render realistic humans. The model also powers Adobe's new prompt based editing features that make specific adjustments based on descriptions, and a new layered image editing tool coming to Photoshop that makes precise contextual changes, such as ensuring shadows are automatically corrected when moving objects around an image for video editors. Adobe also has a new tool for its Premiere Pro software that automatically stencils out people and objects in video frames. AI Object Mask is available in public beta and aims to make it easier to quickly color grade, blur and add visual effects to moving backgrounds by removing the need to manually mask subjects using the pen tool. End quote finally today, the Journal takes a look at Anthropic's alleged success in corporate AI, which apparently makes up around 80% of its revenue. A July report put Anthropic's enterprise API market share at 32%, which would be above OpenAI's reported 25% quote. OpenAI recently inked hundreds of billions of dollars of deals to build data centers filled with chips it hopes will further its AI dominance. But one of its rivals, the Amazon backed developer Anthropic, has a clearer path to making a sustainable business out of AI. Anthropic and OpenAI do similar things. They develop advanced AI models upon which chatbots implement image generators and a host of other AI tools are based. But they have approached the question of how to generate revenue and, one would hope, profit from AI in different ways. Outside of OpenAI's close partnership with Microsoft, which integrates OpenAI's models into Microsoft software products, OpenAI mostly caters to the mass market. Its user base is in large part replacing search engine queries with bot conversations, which has proved immensely popular. ChatGPT had more than 800 million weekly users as of this month, according to the company, which has helped OpenAI reach an annual revenue run rate of around $13 billion, around 30% of which, it says comes from businesses. Anthropic has generated much less mass market appeal. The company has said about 80% of its revenue comes from corporate customers. Last month it said it had some 300,000 of them. That focus has helped put anthropic ahead of OpenAI among business users. Its cutting edge CLAUDE language models have been praised for their aptitude encoding. A July report from Menlo Ventures, which has invested in Anthropic, estimated via a survey that anthropic had a 42% market share for coding compared with OpenAI's 21%. Anthropic is also now ahead of OpenAI in market share for overarching corporate AI use. Menlo Ventures estimate it at 32% to OpenAI's 25%. Anthropic is also surprisingly close to OpenAI when it comes to revenue. The company is already at a $7 billion annual run rate and expects to get to $9 billion by the end of the year, a big lead over its better known rival in revenue per user. Both companies have backing in the form of investments from big Companies Microsoft for OpenAI and a combination of Amazon and Google for Anthropic that help provide AI computing infrastructure and expose their products to a broad set of customers. But Anthropic's growth path is a lot easier to understand than OpenAI's corporate customers are devising a plethora of money saving uses for AI in areas like coding, drafting legal documents and expediting billing. Those uses are likely to expand in the future and draw more customers to Anthropic, especially as the return on investment for them becomes easier to measure. Demonstrating how much demand there is for Anthropic among corporate customers, Microsoft in September said Anthropic's leading language model, Claude, would be offered within its Copilot suite of software. Despite Microsoft's ties to OpenAI, the mass market consumer revenue model is more nebulous. OpenAI has yet to settle on a way to make money from it beyond charging subscription fees. It has a $20 a month plus plan and a $200 a month pro plan for consumers, in addition to a free tier that comes with limits on queries and runs more slowly. Such subscription fees aren't enough to offset the massive cost of developing and rolling out cutting edge AI. The obvious revenue stream for OpenAI's consumer business will be advertising, but it isn't clear how OpenAI or its competitors would inject ads into chatbots that won't be as straightforward as search ads. Users wouldn't likely welcome brand placement in their bot chats. And as it looks for an AD revenue model, OpenAI is in the unenviable position of competing with Google, which has its own suite of mass market AI tools and far deeper roots in advertising. Of course, OpenAI is making a strong appeal to business customers too, both through Microsoft and on its own. And there is an argument that OpenAI's vast user base and exposure to a wider set of queries will give it an edge among corporate users. Yet there is also a possibility that OpenAI's mass market appeal becomes a turn off for corporate customers who want AI to be more boring and useful than fun and edgy. OpenAI recently said it would begin allowing adults to have erotic conversations with ChatGPT and has urged a hands off approach to AI regulation. Even if the company makes its products more constrained in corporate contexts, its freewheeling reputation seems likely to limit its inroads for all of OpenAI's spending. Meanwhile, anthropic has shown it is as good or better in AI arenas that companies care about. Val's AI, a startup that evaluates AI models, ranks the latest version of Anthropic's large language model Claude as top in a business focused benchmark that brings together finance, legal and coding tasks. Anthropic is laser focused on these agentic enterprise use cases and they're playing a very competitive game with OpenAI right now, said Rayan Krishnan, a co founder of Vals. End quote. Hey everybody, the last link in the show notes today is to that interview with Wikipedia founder Jimmy Wales that I've been promising you. There's also a link to an interview with Margit Venmachers, the secret power behind a 16Z, and also a link to an interview with venture capital legend Fred Wilson. That's because the Internet History Podcast is back Search your podcast apps now and subscribe. Some of you have been listening to me since the very beginning, but for those that don't know, the Internet History Podcast is what first got me into podcasting 12 years ago. Now it's what got me my TED residency. It's what led to my book. It's what led me to this show. It's been on hiatus for about six years, but I've banked about 20 interviews to bring it back with, and I'm relaunching it today with those three episodes I've linked to in the show notes. So please subscribe to the Internet History Podcast in your app of choice. But of course, these days podcasts are videos too, right? So subscribe to the Internet history podcast on YouTube. YouTube. It's crazy. Video was not an option when I started podcasting 12 years ago. I had to use Skype and I often had to walk people through how to even use that. Much, much easier now to have a video of each interview. And 12 years ago nobody was doing business history, much less tech history. People kept saying is what I did interesting or old enough to be history? There's a whole industry of this stuff now. Very, very Bottom link in the show notes is to a blog post announcing the podcast's return. If someone would be so kind as to post that to Hacker News or Reddit or wherever, I'd be much obliged. Anyway, I'm happy I finally found a way to fit the Internet History podcast back into my schedule. So subscribe and get ready for the thing that I do best. Tech history Brian style.
- [21:14]
B
What the hell is going on right now and why is it happening like this? At Wired, we're obsessed with getting to the bottom of those questions on a daily basis. And maybe you are too. I'm Katie Drummond, the global Editorial Director of Wired, and I'm hosting our new podcast series, the Big Interview. Each week I'll sit down with some of the most interesting, provocative and influential people who are shaping our right now. Big Interview conversations are fun.
- [21:41]
A
I want a shark that that eats.
- [21:43]
B
The Internet, that turns it all off, unfiltered and unafraid.
- [21:47]
A
So in a lot of ways, I try to be an antidote to the unimaginable faucet of reactionary content that you see online. To the best of my ability, every.
- [21:55]
B
Week we're going to offer you the ultimate luxury of our times, meaning and context. True or false? You, Brian Johnson, the man sitting across from me, one day, at some point, as of yet undefined in the future, you will die. False. Tell me more. Listen to the Big Interview right now in the same place you find WIRED's Uncanny Valley podcast. Subscribe or follow wherever you get your podcasts.