← Tech Brew Ride Home
Tech Brew Ride Home
AI Superapps
00:20:21
OpenAI is becoming a superapp. Does Amazon really want to try its hand at a smartphone again? Google is making further steps to obviate classic Google Search in that innovator’s dilemma way. And in the Longreads, why has AI gotten worse at writing.
Loading summary
Transcript8 lines
- [00:02]
Sponsor/Ad Reader
This episode is brought to you by focus features. On March 27, Focus Features invites you to be a part of the most explosive movie of this year's Sundance and South by Southwest film festivals. The AI Doc, or How I Became an Apocalyptimist is being called supremely entertaining and the most urgent movie of our time. The AI doc, or How I Became an Apocalyptomist, rated PG13 only in theaters March 27.
- [00:34]
Brian McCullough
Welcome to the Tech Brew Ride home for Friday, March 20, 2026. I'm Brian McCullough. Today, ChatGPT is becoming a super app Does Amazon really want to try its hand at a smartphone again? Google is making further steps to obviate classic Google search in that innovator's dilemma way and in the long reads. Why has AI gotten worse at writing? Here's what you missed today in the world of tech. As I said yesterday, the whole AI race has reconfigured itself the first quarter of this year to become a chase for stuff like this. OpenAI plans to merge ChatGPT, Codex and its browser into a desktop super app to simplify the user experience and focus on engineering and business customers. Great. Quoting the Journal the strategy change marks a major shift from last year, when OpenAI launched a series of standalone products that didn't always resonate with users and sometimes created a lack of focus within the company. OpenAI executives are hoping that unifying its products under one app will allow it to streamline resources as it seeks to beat back the success of its rival anthropic. OpenAI is seeking to focus on creating so called agentic AI capabilities within the new super app, in which artificial intelligence systems can work autonomously on a user's computer to carry out a variety of tasks, including writing software and analyzing data, according to OpenAI. We realized we were spreading our efforts across too many apps and stacks and that we need to simplify our efforts, Fiji Simo shared in an internal note with employees Thursday that fragmentation has been slowing us down and making it harder to hit the quality bar we want. Top executives, including Chief Executive Sam Altman, Chief Research Officer Mark Chen and Simo have spent the last few weeks reviewing OpenAI's product portfolio and looking at areas to deprioritize, the Wall Street Journal reported. In an all hands meeting last week. Simo employees they couldn't afford to be distracted by so called side quests. Given Anthropic's rapid success winning over enterprise and coding customers, an OpenAI spokeswoman said the company was very much acting as if it were under a code red, an OpenAI spokeswoman said. The new super app will enable teams inside OpenAI to work more closely together and help the research division focus its efforts around improving one central product. Over the coming months, the company expects to add new agentic capabilities within its Codex app so it can help with productivity related tasks beyond coding before merging ChatGPT and the Atlas browser into the super app as well. The mobile ChatGPT app will remain unchanged. OpenAI's organizational structure grew complicated due in part to the myriad products that it announced last year, including its video generator Sora and a new hardware device. This is an opportunity to combine the strongest AI consumer app and brand with the strongest agentic app and really leverage our consumer scale to give agentic capabilities to everyone, simos said in her note. End quote. The White House has released an AI policy framework explicitly calling on Congress to preempt state AI laws, create age gating requirements for AI models, and more. Quoting Politico. The light touch framework blends the Trump administration's effort to create a national AI rulebook on issues like political bias within models and reducing barriers to innovation with protections for children and teens online. It also urges Congress to overrule state AI laws that the administration says impose undue burdens in favor of the minimally burdensome federal law that it's recommending. The Trump administration has been trying to establish preemption over state AI laws using Congress and executive orders for roughly a year, arguing that the patchwork of laws harms AI innovation. The framework explicitly calls on Congress to preempt any state laws that regulate the way models are developed or that penalize companies for the way their AI is used by others, and instructs US Lawmakers not to create any new federal agencies to regulate AI. The proposal outlines some areas where the federal government's laws wouldn't overrule those of the states and ask Congress to allow states to keep laws that protect children, including those that ban AI generated child sexual abuse material. The framework also asks Congress to create age gating requirements for models likely to be accessed by children and to give parents tools to set up safeguards around their children's use. It does not go as far as some Republicans have called for, such as proposals to roll back liability shields for tech companies. In addition, it calls on federal lawmakers to pass legislation that encourages AI skills training, education as well as data collection on job disruption that stems from AI. The White House also recommends in the document that Congress codified Trump's ratepayer protection pledge signed by companies including Amazon, Google and OpenAI earlier this month, requiring tech firms to supply or pay for the electricity used by the data centers they operate. End quote. Well time for another bite of the Apple I guess. See what I did there? Sources tell Reuters that Amazon's Zero1 unit is developing Transformer, a phone that syncs with Alexa. It would be Amazon's second attempt at a smartphone after the infamous fire phone in 2014, quote the latest effort, known internally as Transformer, is being developed within its Devices and Services unit, according to four people familiar with the matter. The phone is seen as a potential mobile personalization device that can sync with home voice assistant Alexa and serve as a conduit to Amazon customers throughout the day, the people said. The initiative is the newest chapter in a years long effort to bring to market Jeff Bezos long held vision of a ubiquitous voice driven computing assistant. Akin to the voice controlled computer in science fiction series Star Trek, Bezos had envisioned a smartphone that had shopping at its core and could take on Apple by offering shipping, conveniences and discounts through the prime membership program. Along the way, Amazon could gain a wealth of new data about users only available through mobile phones, combined with purchase history and content content preferences. Amazon's effort to develop a new smartphone has not been previously reported. Reuters could not determine some details such as the anticipated price of the phone, the revenue Amazon hopes to generate or the financial commitment Amazon has made to the project. End quote. US Prosecutors have charged three people affiliated with Super Micro, including co founder Yi Xian Liao, with smuggling Nvidia AI chips into China. Super Micro stock dropped more than 25% on the news. Quoting CNBC in an indictment unsealed Thursday, the US Government alleged that Yixian Wali Liao, Ryu Tsian, Steven Chang and Ting Wai Willi sun worked together to violate the Export Control Reform Act. The server company's products containing Nvidia chips, quote, are subject to strict US Export controls barring their sale to China without a license. The plaintiff said in the indictment those controls are in place to protect US national security and foreign policy interests, among other things. Liao is a co founder of server maker Super Microcomputer and a member of its board of directors. He controls $464 million worth of Supermicro shares, according to FactSet. He did not respond to a request for comment. Shares of Supermicro fell 25% on Friday after a federal court released the indictment. Supermicro said that while the company isn't named as a defendant, Liao works as senior vice president of Business Development. Chang is a sales manager in Taiwan and Sun is a contractor. The company has placed the employees on leave and ended its relationship with the contractor. Liu and sun were both arrested Thursday, while Chang is a fugitive, the attorney's office said. End quote. Sources say Jeff Bezos is in talks to raise $100 billion for a fund that would buy companies in industrial sectors like chipmaking and defense and automate them with AI. Quoting the journal the Amazon.com founder is meeting with some of the world's largest asset managers to raise funding for the project. A few months ago, he traveled to the Middle east to discuss the new fund with sovereign wealth representatives in the region. More recently, he went to Singapore to raise funding for the effort as well, according to people familiar with the matter. The fund, described in investor documents as a manufacturing transformation vehicle, is aiming to buy companies in major industrial sectors such as chipmaking, defense and aerospace. It would dwarf the size of some of the world's largest buyout Funds and rival SoftBank's $100 billion tech focused vision Fund. While much of the AI revolution has been focused on large language models, billions of dollars have begun to flow to companies that are seeking to apply spatially focused AI systems toward industries including robotics and manufacturing. Language based AI models are being used to automate software engineering, and leading companies are seeking to boost their use in ways that can also affect knowledge work for finance, real estate and other industries. A broad push toward AI infused automation is also moving through manufacturing related work, although startups and companies focused on this area are in an earlier stage. While it's unclear the extent to which automation could impact jobs in those industries, tech and E commerce fulfillment companies have been applying that technology in warehouses for years. Amazon, one of the country's largest employers, has closed in on the milestone of having as many robots as humans in its workforce. End quote.
- [09:57]
Sponsor/Ad Reader
So good, so good, so good. Spring styles are at Nordstrom Rack stores now and they're up to 60% off. Stock up and save on Rag and Bone, Madewell, Vince, All Saints and more of your favorites.
- [10:09]
Brian McCullough
How did I not know?
- [10:10]
Sponsor/Ad Reader
Rack has a why do we rack for the hottest deals? Just so many good brands. Join the NordicLub to unlock exclusive discounts, shop new arrivals first and more. Plus buy online and pick up at your favorite Rack store for free. Great brands, great prices. That's why you rack
- [10:28]
Brian McCullough
bad news for small teams. You're still a target for cybercriminals who know lean teams often lack the resources to prevent or respond to a breach. The good news is your team regardless of size, can foil cybercrime with 1Password. They provide simple security to help small teams manage the number one risk that bad actors weak passwords. Take the first step to better security by securing your team's credentials. Find out more at 1Password.com Ride so you can secure every login that's 1Password.com Ride
- [11:04]
Sponsor/Ad Reader
it's crunch time at work and you need to bring wings to your workday. Visit redbull.com gettingitdone and answer a couple questions about your work style to get a Spotify customized player playlist tuned to your productivity. Plus, score a can of Red Bull on us while you go from to do to done. And remember, Red Bull gives you wings. Supplies are limited. Terms apply. Visit the website for more information. Score more with the college branded Venmo debit card and earn up to 5% cash back with Venmo Stash Got paid back with the Venmo debit card, you can instantly access your balance and spend on what you want, like game day, snacks, gear, tickets and more. The more you do, the more cash back you can. Plus there's no monthly fee or minimum balance. Sign up now@venmo.com collegecard the Venmo Mastercard is issued by the Bancorp Bank NA Select Schools available Venmo Stash terms and exclusions apply at venmo me stashterms max $100 cash back per month Google is
- [12:00]
Brian McCullough
running what it calls a small experiment, replacing news headlines in search results with AI generated ones after adding the feature in Google Discover in January, quoting the Verge. Since roughly the turn of the millennium, Google search has been the bedrock of the web. People loved Google's trustworthy 10 blue links search experience, and its unspoken promise of the website you click is the website you get now. Google is beginning to replace news headlines in its search results with ones that are AI generated after doing something similar in its Google Discover news feed, it's starting to mess with headlines in the traditional 10 blue links, too. We found multiple examples where Google replaced headlines we wrote with ones we did not, sometimes changing their meaning in the process. For example, Google reduced our headline I used the Cheat on Everything AI tool and it didn't help me cheat on anything to just five words Cheat on everything AI tool. It sounds almost like we're endorsing a product we do not recommend at all. What we are seeing is a small and quote narrow experiment, one that's not yet approved for a fuller launch, Google spokesperson Jennifer Cutts, Mallory De Leon and Ned Adriance tell the Verge they would not say how small that experiment actually is. Over the past few months, multiple Verge staffers have seen examples of headlines that we never wrote appear in Google search results, headlines that do not follow our editorial style and without any indication that Google replaced the words we chose. And Google says it's tweaking how other websites show up in search too, not just news. The good news for now is that these changed headlines seem to be few and far between, and they're not yet the kind of tripe we've seen in Google Discover. But these are just the first headlines we've seen Google change. They may be the canary in the coal mine. Google may alter the deal even further. While Google says this is an experiment, you shouldn't assume that means the company won't roll it out more widely, because Google originally told us its AI headlines in Google Discover were an experiment too. A month later it told us those AI headlines are now a feature, one that performs well for user satisfaction. Google did not explain why the company is no longer respecting the headline identifiers it has long encouraged newsrooms to use. The company did answer some specific questions via email, though. Google told us that the overall idea is to quote identify content on a page that would be a useful and relevant title to a user's query. The goal is better matching titles to users queries and facilitating engagement with web content. According to Cuts, this test is quote not specific to news publications, but looking at how we can improve tiles horizontally. According to Adriance, Google confirmed that the test uses generative AI, but claimed that if we were to actually launch something based on the experiment, it would not be using a generative model and we would not be creating headlines with Gen AI. According to De Leon, Google did not explain how it might replace our story titles without using generative AI. Mostly, Google's answers tried to normalize the idea of replacing headlines in search, suggesting that this is just one of the tens of thousands of live traffic experiments that Google runs to test possible improvements to Google search and reminding us that it's already been tweaking the titles of web pages in search to help users for many years now. But I want to be clear, this is not normal. I've edited tech news for 15 years, paying close attention to SEO, and I've never before seen Google overwrite a headline in search results with something it created itself. The changes that Google typically makes to a news story's title are far simpler. If Google's algorithms decide a headline is too long or lopsided, it'll sometimes show you only part of that headline lopping off the beginning or the End? In the longreads this week, you might have seen that story making the rounds on social media about someone using ChatGPT to cure a dog's cancer. I'm sorry to say, but quoting the Verge, A few weeks after Rosie's first injection last December, Cunningham said her tumors had shrunk and she's doing better, even chasing rabbits in the park. They've not disappeared entirely, though, and one tumor didn't respond at all. I'm under no illusion that this is a cure, but I do believe this treatment has bought Rosie significantly more time and quality of life, cunningham told the Australian. That nuance was lost as the story spread. Newsweek ran the headline Owner with no medical background invents cure for dog's terminal Cancer, while the New York Post declared that a tech pro saves his dying dog by using ChatGPT to code a custom cancer vaccine. On social media, many accounts hyped Rosie's case as a cure and a sign a new era of personalized medicine had arrived. Some, notably OpenAI President and Co founder Greg Brockman, should have definitely known better. And others, like Google DeepMind CEO Demise Hassibas, did and shared it without hype. Elon Musk joined in, too, keen to point out that Xai's Grok also played a part, a detail that was absent from much of the original coverage. The story also gives AI far too much credit. Not only was Rosie not cured of cancer, it's not clear the MRNA vaccine was responsible for her improvement in the first place. End quote Then, from the Atlantic Is it just me or has AI actually gotten worse at writing, you know, obvious AI in text and stuff? Quote I talked with people who would know people who work at LLM companies, AI data vendors, academic computer science departments, and AI writing startups. Some spoke with me under the condition of anonymity because their employers barred them from speaking publicly about their work. What I learned is that modern LLMs are built in a way that is antagonistic to great writing. They are engineered to be rule following teachers, pets that always have the right answer in hand. In many respects, they've come a long way from GPT2, but they've also lost something that made them looser and more compelling. End quote. No bonus content for you on this feed this weekend, but as soon as I hit publish here, I'm gonna go edit and hit publish on episode two of the series on the Cold War that I'm doing. Check that out over at the RAD History feed. Talk to you on Monday.