Tech Brew Ride Home - Omnibus Week Of 04/28

Episode Date: May 3, 2025

Omnibus Week Of 04/28 Learn more about your ad choices. Visit megaphone.fm/adchoices

On April 4th, 2023, around 2 in the morning, a man was found stabbed multiple times on a sidewalk in downtown San Francisco. Hey, who did this to you? What happened next turned the story into a political firestorm. Reports have identified the victim as Bob Lee, the founder of Cash App. From Bloomberg Podcasts, this is Foundering, the Killing of Bob Lee, beginning April 16. This is the Omnibus episode for the final week of April 1st week of May 2025. Mark German, Apple Scoop Monday, back to the manufacturing iPhones in India thing. Turns out there might be one big fly in the ointment for Apple.

The big plans that they have for the 20th anniversary iPhone, quote, as of right now, Apple has enough capacity in India to meet roughly a third of annual U.S. demand. Upcoming factories in India, including one intended to be the second biggest iPhone, plant in the world would help Apple get the rest of the way there. This past week, we reported that Apple is aiming to produce the bulk of its U.S. iPhones in India by 2027. While I do ultimately expect this to happen, it's far from a sure thing. There is no telling how the tariff policies will change tomorrow, let alone 18 months from now. But even if Apple does move the bulk of U.S.-bound iPhone manufacturing to India, it almost certainly won't cover everything. Apple is working on two

major new iPhone models for the device's 20-year anniversary. The original model was announced at the start of 2007 and released later in the year, so there's a little wiggle room around the timing of its birthday. That includes the company's first foldable and a new more glass-centric pro model. The idea that Apple would want to build those devices outside of its China comfort zone is slim, at least at the outset. While Apple's manufacturing in India has reached parity with China in terms of current iPhone quality, the 20th year anniversary models are extraordinarily complex. They'll require new parts and production techniques, making it far from a certainty that Apple will be able to build those outside of China. At some point, yes, but certainly not by the year 2027. Keep in mind, Apple has never produced a major new

product design outside China for the first go-round, end quote. Actually, to this end, the FT has a deep dive into Apple's iPhone manufacturing process, which uses around 2,700 parts from 187 suppliers in 28 countries. And what they found was that just 30 of those suppliers had zero presence in China. The rest of them all did. For example, the frame, the boring old aluminum frame, because the frame is made from a single piece of aluminum, a metal block is cut and shaped, it uses highly specialized machines to do that. that only available at scale in China. Quote,

The close proximity of suppliers and manufacturers is crucial to Apple's productivity. There are a lot of advantages to co-locating the activities in the supply chain in terms of speed and quality of communication and innovation in the product and process design, says, say, the professor at Santa Clara's Levy School of Business. It means that you can get deliveries very quickly and you can communicate with your supplier very easily. And when you put an ocean between the customer, in this case Apple and the component supplier, there is a disadvantage, he adds. This electronic ecosystem is why moving assembly to the U.S.,

quote, introduces inefficiencies, according to Wamsi Mohan at Bank of America. If everything is not made close by, it gets complicated, he said. And while Apple may be able to find alternative manufacturers for some iPhone components, some are sole sourced. Taiwan's TSMC, the world's largest chipmaker supplies the main processor. Although the company says mass production began in Arizona in January, experts say there is no replacement for the chips produced in Taiwan and South Korea, end quote. So really, this was ostensibly a story about the iPhone, but it's about tariffs in the end, right? And even if the iPhone gets exemptions, as we've asked before, what if you're not Apple? The Wall Street Journal has a piece looking at how, while big tech is largely spared from tariffs

so far, small companies making devices like, I don't know, baby monitors, face price hikes, shortages and degrading product quality. Quote, on May 2nd, the de minimis trade exemption for Chinese imports whose retail value is $800 or less will end. Some products will see immediate price surges. Lofty, which makes a smart sunrise lamp, has about 750 units left in stock. The Made in China lamp now carries tariffs totaling 175%. On May 1st, Lofty's sticker price rises from $275 to $450.

$100 of that is to the U.S. government, says Chief Executive Matthew Hassett. Floss, an electric flosser, has about eight weeks of inventory left in the U.S. Another two and a half months worth is on hold in China, where the manufacturer is getting antsy for payment. Floss CEO Samantha Cox says China was the best option to manufacture the $119 gadget. The country specializes in the precision assembly required to install the flosser's tiny motor, battery, and other components, she says. Cox plans to store the inventory in bonded warehouses in the U.S. secure facilities that allow importers to defer customs payments. If tariffs go down, her product is ready to sell. No one's margins can take a 145% tariff, says Cox. Foss will only raise prices as a last resort,

but will remove perks such as free shipping. Her immediate concern is whether the current stock will last until Amazon's prime day in mid-July when there's typically a price markdown. That's a huge event for a product like floss, she says. Should the tariffs stay high, there may be more empty shelves in a few months' time. Without a long-term policy in place, there will be no Christmas, says Grant Priggy, CEO of Smart Air Purifier Maker Mila. Everything you're buying today was imported pre-tariffs, but those warehouses will run out in the next 30 to 90 days, he says. Even if China tariffs end up lower, there will also be a massive increase in companies' shipping costs. Prigy says the 90-day pause on levies in countries other than China prompted a rush in shipments.

Boat prices from Vietnam, where Mila is moving some manufacturing, went up 25% last week, he says, end quote. Not tariff news, but it is news if you were an early adopter. Google says it will end support for first and second generation nest thermostats on October 25th and won't launch new models in Europe due to its, quote, unique heating systems. Quoting the Verge. Google has just announced that it's ending software updates for the first-generation Nest learning thermostat released in 2011 and the second-gen model that came a year later. This decision also affects the European Nest Learning Thermostat from 2014.

You will no longer be able to control them remotely from your phone or with Google Assistant, but can still adjust the temperature and modify schedules directly on the thermostat the company wrote in a Friday blog post. The cutoff date for software updates and general support within the Google Home and Nest Apps is October 20, In other significant news, Google is flatly stating that it has no plans to release additional nest thermostats in Europe. Heating systems in Europe are unique and have a variety of hardware and software requirements that make it challenging to build for the diverse set of homes. The company said, the Nest Learning Thermostat 3rd Gen 2015 and Nest Thermostat E 2018 will continue to be sold in Europe while current supplies last, end quote. Losing the ability to control these smart

thermostats from a phone will inevitably frustrate customers who've had Nest hardware in their home for many years now. Google's not breaking their core functionality, but a lot of the appeal and convenience will disappear as software support winds down. The early Nest learning thermostats can at least be used locally without Wi-Fi, which isn't true of newer models. There's one bright spot for owners of recent Nest thermostats, though. Google says owners will be able to create and adjust schedules from the Google Home app later this year for the first time. Still, this type of phase out is a very real fear tied to smart home devices as companies put screens into more and more appliances. Is 14 years a reasonable lifespan for these gadgets before the smarts fade away?

There's no indication that Google plans to open source the hardware. In a clear attempt to ease customer anger, Google is offering a $130 discount on the 4th Gen Nest Learning Thermostat in the U.S., $160 off the same device in Canada and 50% savings on the Tadu Smart Thermostat X in Europe since the NEST lineup will be gone soon, end quote. They broke out of the gate with buzz like they might be a deep-seek mock two, but then we kind of haven't heard much from them, or at least haven't reached my radar. Nonetheless, sources say Chinese AI startup Manus raised $75 million, led by benchmark at a $500 million valuation and plans to expand to markets, including the U.S., Japan, and the Middle East,

quoting Bloomberg. Manus in March previewed what it called a general AI agent capable of screening resumes, creating trip itineries, and analyzing stocks in response to basic instructions. Its service performed better on some fronts than OpenAI's deep research. Another recently released agent, the company claimed at the time, since then several companies from Bight Dance to Bidu have followed suit with their own competing agentic AI platforms. While AI agents require a certain amount of hand-holding Manus, co-founder, and chief scientist

G. E. Chow, has said its product is truly autonomous. A slick video demonstration from the company quickly went viral, prompting a scramble for a limited number of access cores and earning the startup comparisons to Deepseek, the Chinese startup that rattled Silicon Valley in January with a capable yet low-cost model. Like Deepseek, Manus sparked questions about the U.S. lead on artificial intelligence, this time in a product category that American tech companies see as a key investment area. Initial reactions from Manus users were mixed. Some declared it groundbreaking. Others said it felt half-finished. Last month, the startup began offering a $39 per month subscription and a $199 upgraded option

in line with OpenAIs chat GPT Pro, though somewhat aggressive pricing for a membership service still in a testing phase, end quote. Speaking of potentially fading excitement about something, the journal says that prompt engineering roles, one of the busiest jobs back in 2023 when the AI moment first supernobod, have become obsolete as AI models, better intuit user intent, and companies train staff in writing prompts. Quote, two years ago everybody said, oh, I think prompt engineer is going to be the hot job, said Jared Spataro, chief marketing officer of AI at work for Microsoft. It's not turning out to be true at all. As part of a recent survey commissioned by Microsoft, 31,000 workers across 31 countries were asked

what new roles their companies were considering adding in the next 12 to 18 months. Prompt engineering was second from the bottom, Spataro said, while roles such as AI trainer, AI data specialists, and AI security specialists top the list. Spataro said large language models have evolved to be more iterative, conversational, and aware of context. Microsoft's AI-powered research agent, for example, will ask follow-up questions, tell you when it doesn't understand things and ask for feedback on the information it serves up, Spittaro says. In other words, you don't have to have the perfect prompt. On job platform,

Indeed, the number of postings for prompt engineers is minimal, said Indies VP of AI, Hannah Calhoun. User searches on Indeed for the role surged from two searches per million total searches in the U.S. in January 2023, months after ChatGPT's debut to 144 per million in April 2020. They have since flatlined to about 20 to 30 searches per million, according to Indeed. People were certainly interested in the role at the time, Calhoun said, but that interest hasn't been matched by actual employer job postings. Maybe they talked about the value of prompt engineers, but they weren't then actually hiring for that, said Calhoun. Squeezed by tight budgets and growing uncertainty with the economy, companies have been much more measured in their overall hiring in recent years.

Companies such as insurer nationwide, workwear brand Carhart, and insurance company New York life all said they've never hired prompt engineers, but instead found that to the extent better prompting skills are needed, it was an expertise that all existing employees could be trained on. Nationwide, for example, has rolled out a company-wide AI training program for all employees, with prompt engineering being but one of the most popular courses within it, said, Chief Technology Officer Jim Fowler. So whether you're in finance, HR, or legal, we see this becoming a capability within a job title, not a job title to itself, he said, end quote. Finally today, top of tech meme for the last 24 hours has been a story from Ben Smith at Semaphore,

taking a look at the network of group chats that surged in 2020 and have only grown over the last several years as a place where Silicon Valley Insiders meet and chat, often revolving around Mark Andresen, but also podcaster Eric Torrenberg, who, by the way, his podcast was just last week, sort of aquired by A16Z. This is one of those stories that I have. kind of debated covering because is it too inside baseball? Is it maybe too gossipy rather than actually newsy? I have a link to the piece at the end of the show notes today. So read that if you want the goss. But since Ben Smith is a friend of the pod, I thought the way I'd cover it would be to get Ben on the phone real quick and ask him about it. Ben Smith, thanks for popping on the pod real quick. Yes, fun to pop. Thank you for having me. So I know you're not going to

tell me who it was that gave you access to the various group chats that you mentioned in the piece. But I'm curious what you thought the motivation was for people to share with you the existence of chats like this? Well, you know, I had a couple of leaked screenshots. Honestly, not that much because they use disappearing messages. And I, but I was actually surprised how many people were willing to talk to me on the record or kind of in great detail on background about these chats, you know, that have been a huge, but it's really, I think it's been a huge part of the lives of lots of powerful people and not powerful people who are, you know, shaping American culture over the

last five years. And I think people feel proud about them. They feel honestly, like really sentimental. Like, I think everybody realizes that it's an era that's sort of coming to a close, partly as Trump really divides this community. And I think a lot of them look back, you know, this having been, I mean, one of them described it to me as the Republic of Letters. A lot of them invoke Parisian salons. I mean, I do think the, um, Parisian salons were probably more fun. Well, that's kind of what I wanted to talk about. You mentioned journalist, which was sort of like a left wing, sort of news served from like 15 years ago.

Central left. Right, right. And so one of the things that I was thinking about is these things never stay secret forever. But also, and I'm going to push back a bit, I kind of feel like these things are everywhere. I don't know that this is coming to an end. Like, in a way, you could say this is sort of like that astronaut always has been meme. But I find it, and I find it funny how surprised people in North America are that these group chats exist. Because I feel like in Europe, they've had entire governments brought down by group chat scandals for years.

But I feel like this is the mode for powerful people to commiserate these days. Oh, for sure. I don't mean the group chats or group chats among powerful people coming from this specific network of sort of center-right, anti-woke, pro-Trump. Trump chats is coming to an end because about half the people in them are no longer pro Trump and are like, wait a second, how did I get into this kind of group think world where I got it, you know, and so there's these bitter, bitter disagreements now, which are not where, you know, yeah, so the group think is sort of shattering. No, I mean, I think that there have always been these kind of inside conversations, whether it was like the boys on the bus or letters in the 17th century

or group chats. Now, I do think disappearing messages, I do think real privacy makes a difference, actually. Like, I think we're, most so interesting of doing this story is that virtually none of this is preserved. And I think that's basically a good thing. I mean, it's, you know, spoken,

it has a little more the character of spoken conversation than this sort of weird era we've had in which every stray thought has been, you know, documented forever. Right. And, like, that's what I'm saying. Like, if somebody listening to this, your CEO has a group chat that you don't know about.

Like, we've heard whispers for years that, like, the C-suite at major tech companies have completely exclusive lockdown messaging systems. I even remember interviewing somebody about a company once where they're, like, all of the real stuff goes down on, like, a CC email thread that's been running for years because the CEO only uses email to do work, you know? Totally. But, like, the idea that Signal and these sorts of group chats, like, it's not just the Signalgate stuff. Like this is where increasingly like a level of powerful or famous person, they feel they need this sort of space for themselves.

Yeah, but yes, that's certainly true. And Signalgate was just a reflection of the fact that they're all in Signal all the time, shooting emojis at each other. But I mean, I do think that that sort of period after 2020 when social media was really hot and very progressive chased a lot of people who, who disagreed and were scared to disagree into more private spaces. But one of the people who's in these big group chat, one of the big group chat said to me that was so interesting was that now the thing that you're afraid to express online is criticism of Donald Trump.

And that the people who are critical of Trump in the group chats are keeping that in the group chat because they don't want professional or other consequences for criticizing Trump. One of the things that I think Silicon Valley folks will recognize reading between the lines of your piece is sort of this is where things come from like you know I'm thinking about like group chats for VCs and stuff like that especially investing as a very group think sort of activity but like I remember that whole Silicon Valley Bank weekend like if yes

looking at it from the outside it's like how come somebody everybody always suddenly woke up and was on the same page freaking out it's because of chats like this like where it is seem from the outside Lex Freeman or Dworkish Patel seemingly comes out of nowhere, but suddenly can get interviews with the most powerful people in the valley. It's these group chats. It's sort of this community. They literally were the vehicle for a bank run, which is the ultimate kind of herd mentality group think thing. And I think that's, you know, to me that's the danger of them, that you can really wind up in these bubbles and also wind up with this feeling. And I was in journalists. And, you know, I was, I'm kind of of the center, not, wasn't really partisan, but felt it a little

We're like, oh, these are my friends and they're on my team. And so we're going to work out our disagreements, and we can have honest and open disagreements inside. But in the outside, we are, you know, and I never felt this exactly, but felt the poll. We are rapid partisans who defend each other to the death. And actually, Balaji, Srinivasan, and Joe Lonsdale had a totally normal argument that I quote in the piece.

And when I reached out to them for comment, they went bananas on Twitter and on X. and swore that they were friends forever and no disagreement in private could possibly tear them apart in public, which was kind of symptomatic of that. Like, wait, you're not allowed to have, like, everything you say in public is just a, you're just sort of out there brigading, and the things you say in private are real. Like, I don't think that's super healthy. Well, that's the other thing that I wanted to point out, like, reading between the lines. It's sort of like, this is how social media operates for the other half. Like, you talk about these group chats, like, coming and going.

Like instead of social media where you have like this public history and this audience that follows you around and like you have content and takes for them. But then for these private chats, it's sort of like that Stefan sketch from S&L. Like when you get invited to a new one of these, it's like, oh, the cool kids have found me again. And it's like really active for a few months. But then it can kind of fade. And then you get invited to a new one. And it's almost like changing personas or like changing high schools. So that it's social media where you can reinvent yourself when you get that.

in the next sort of cool kid group like this. Totally. And it has all of the sort of seductions of this kind of access to power. Like lots of my friends and sources who run them, I sure for years would be like, I can't believe this, but like Mark Andresen is blowing out my phone. Like that is kind of cool. But there is also, I mean, something I have found in this moment

of this kind of maximal transparency and dysfunction is that used to be if you were a powerful person, you knew things and had secrets that other people didn't know or have. And, um, and, now it's like you get into like the sanctum sanctorum of the people who like behind the final door of the most powerful and important people who know everything and it turns out they're just like talking about some tweet from three days ago that like is the same thing that like the dads at your barbecue were talking about like there's this strange flattening right um and like a lot of what they're doing these group chats is oh my god did you see this tweet and you should tweet that like those are the two main conversations um last question real quick you mentioned srina vassi and Lonsdale and their reaction online. But what's been the other reaction to some of the folks that are on these chats if you've heard?

You know, honestly, I did not expect this. But, like, it's been incredible. I've been very pleased that a lot of them have been texting an email and saying, hey, like, you actually got the vibe of this basically right. So that's always a nice thing to hear after you write a story. Indeed. The story, as I mentioned, is at the bottom of today's show notes. Ben Smith from Semaphore.

You have a great podcast as well, so I wanted to give you a chance to mention Yes, please check out. If you're interested in all things, media kind of writ large, please check out mixed signals from some of four media. A source says that Amazon plans to soon show shoppers how much President Trump's tariffs are adding to the price of each product right next to its total listed price. Quoting Punch Bowl news, Amazon doesn't want to shoulder the blame for the cost of President Donald Trump's trade war, so the e-commerce giant will soon show how much Trump's tariffs are adding to the price of each product, according to a person familiar with the

the shopping site will display how much of an item's cost is derived from tariffs right next to the product's total listed price, end quote. Now, as we discussed before, people are afraid to go first on stuff like this because they don't want to catch strays. Strays like this. A White House spokesperson this morning already said Amazon doing this is, quote, a hostile and political act. So we will see where that goes. But it should be noted that Amazon technically isn't the first to do this. Kickstarter has introduced a tariff manager tool that lets creators add extra charges for backers on projects that were already fully funded to address U.S. tariffs, quoting 404 media. Over the past few weeks, we've been hard at work developing tariff relevant resources to support

our community, from guidance to help creators navigate rapidly changing policies, to tips on shipping logistics, and even information to help backers better understand the challenges creators are facing. Our focus has been supporting you through uncertain times, but we also know that information alone isn't always enough. Kickstarter said in a blog post published last week announcing the tariff manager tool. Built specifically to address the financial challenges posed by U.S. import tariffs, Kickstarter's tariff manager is designed to give creators more control, flexibility, and transparency at one of the most critical phases of your journey, fulfillment, end quote.

This makes sense because Kickstarter is a lot of bespoke hardware products and projects, right? a vast proportion of which likely sources the components of those hardware projects from China. More tariff stuff. Temu is adding what it calls import charges of between 130 and 150 percent, quote, due to recent changes in global trade rules and tariffs. Sheehan has also begun hiking prices but without explicit fees, quoting CNBC. The fees, which began cropping up over the weekend after price hikes went into effect on Friday costs more than the individual products consumers are buying and can more than double the price of a

typical order. For example, a summer dress sold on Temu for $18.47 will cost $44.68 after $26.21 in import charges are added to the bill, a 142% surcharge, a CNBC analysis shows. A child's bathing suit priced at $12.44 will cost shoppers 31.12 when the $18.68-import charge is taken into account, a staggering 150% fee. A handheld vacuum cleaner listed at 1693 now costs $4011 when factoring in an important charge of $21.68, which is roughly 137% markup. Rival discount retailer Sheehan has also hiked prices on its site, but it doesn't appear to be implementing import charges. The company added a banner, at checkout that states tariffs are included in the price you pay. You'll never have to pay extra at delivery.

The moves come after Temu and Sheean warned earlier this month that they would raise their prices after Trump slapped a 145% tariff on many imports from China and vowed to end the de minimis exemption on May 2nd. The widely criticized loophole helped accelerate Temu and Sheean's growth in the U.S. because it allowed most packages to enter the country duty-free as long as the imports were valued under $800. Due to recent changes in global trade rules and tariffs, our operating expenses have gone up. Temu said on its site earlier this month, to keep offering the products you love without compromising on quality, we will be making price adjustments starting April 25, 2025. Temu has sharply slashed its online ad spending in the U.S. since Trump announced sweeping tariffs.

Temu's ranking in Apple's App Store has since plummeted to number 73 after consistently ranking in the top 10, according to censor-tower data. Sheehan is currently at 54 down from 15 last month. Temu shoppers have flooded a Reddit forum with posts decrying the tariff-induced import charges in the days since the company raised prices. In one post titled RIP Temu, it was nice while it lasted. A user wrote that the price of items went flying up on Friday. From shopping like a billionaire to shopping like a peasant in one day,

a user wrote in a separate Reddit post on Saturday, end quote. The U.S. House of Representatives has passed the Take It Down Act to criminalize posting non-consensual intimate imagery, including deep fakes and requiring online platforms to remove them, quoting the Washington Post. The bipartisan Take It Down Act, which passed the Senate unanimously in February, now heads to the desk of President Donald Trump, who is expected to sign it into law. The bill makes it a federal crime to publish non-consensual intimate imagery or N-CII of any person. and requires online platforms to remove such imagery within 48 hours when someone reports it. That would make it the first significant internet law of Trump's second term and the first U.S. law to take aim at the fast-growing problem of NCII. The bill's passage delighted many advocates for survivors and victims of revenge porn and sextortion scams, while some free expression

and privacy advocates say they worry it will be abused. The legislation's passage by a vote of 409 to 2 marks a victory for First Lady Melania Trump, who has championed the bill as part of her B-Best campaign against cyberbullying. The president indicated in March that he plans to sign it and quip that it is a personal boon, quote, because nobody gets treated worse than I do online, end quote. Today's bipartisan passage of the Take It Down Act is a powerful statement that we stand united in protecting the dignity, privacy, and safety of our children, Melania Trump said in a statement, hundreds of AI undress apps that can forge image. images of real people in seconds have proliferated across the internet in recent years,

harnessing the same wave of technology that is powered image generation tools such as Dolly and Mid-Journey. Some of those apps advertise on mainstream social networks such as Meta's Instagram, despite violating those platforms' rules. Among the most common targets are female celebrities, including singer Taylor Swift and comedian Bobby Altoff, both of whom were the subject of sexually explicit AI fakes that went viral on Elon Musk's Social Network X in 2024. The imagery is also often used to harass, intimidate, or embarrass young women and teens. Those victimized have described their efforts to get non-consensual nudes scrubbed from the internet as a nightmarish game of whackamol, end quote.

Alibaba has debuted its Quen III family of openweight hybrid AI reasoning models, including Quen3-235B A-22B, with 235 billion total parameters and 22 billion activated parameters. Quoting TechCrunch, most of the models are, or soon will be available for download under an open license on AI Dev platforms Hugging Face and GitHub. They range in size from 0.6 billion parameters to 235 billion parameters. Parameters roughly correspond to a model's problem-solving skills and models with more parameters generally perform better than those with fewer parameters. According to Alibaba, the Kwen 3 models are hybrid models.

They can take time to reason through complex problems or answer simpler requests quickly. Reasoning enables the models to effectively fact-check themselves, similar to models like OpenAIs O3, but at the cost of higher latency. Some of the models also adopt a mixture of experts or MOE architecture, which can be more computationally efficient for answering questions. MOE breaks down tasks into subtasks and delegates them to smaller specialized expert models. The Quen3 models support 119 languages, Alibaba said, and were trained on a dataset of over 36 trillion tokens. tokens are the raw bits of data that a model processes. One million tokens is equivalent to about 750,000 words. The company said Quen3 was trained on a combination of textbooks, question and answer pairs,

code snippets, AI generation data, and more. These improvements, along with others, greatly boosted Quinn 3's capabilities compared to its predecessor, Quinn 2. Alibaba said none of the Quinn 3 models seem to be head and shoulders above the top-of-the-line recent models like OpenAIs 03 and 04 Mini, but they're strong performers nonetheless. and quote. OpenAI has begun rolling out product recommendations in ChatGTPT for pro plus free and logged out users with buy buttons that link to merchants' websites, quoting Wired. In a pre-launch demo for Wired, Adam Fry, the ChatGPT search product lead at OpenAI demonstrated how the updated

user experience could be used to help people using the tool for product research decide which espresso machine or office chair to buy. The product recommendations shown to prospective shoppers are based on what ChatGPT remembers about a user's preferences as well as product reviews pulled from across the web. Fry says ChatGPT users are already running over a billion web searches per week and that people are using the tool to research a wide breadth of shopping categories like beauty, home goods, and electronics. The product results in ChatGPT for Best Office Chairs, one of Wired's rigorously tested and widely read buying guides, included a link to our reporting in the Sources tab. The new user experience of buying stuff

inside of ChatGPT shares many similarities to Google shopping. In the interfaces of both, when you click on the image of a budget office chair that tickles your fancy, multiple retailers like Amazon and Walmart are listed on the right side of the screen with buttons for completing the purchase. There is one major difference between shopping through ChatGPT versus Google. For now, the results you see in OpenAI searches are not paid placements but organic results. They are not ads, says Fry. They are not sponsored. So how does ChatGPT choose which products to recommend? Why were those specific espresso machines and office chairs listed first when the user typed the prompt? It's not looking for specific signals that are in some

algorithm, says Frye. According to him, this will be a shopping experience that's more personalized and conversational rather than keyword focused. It's trying to understand how people are reviewing this, how people are talking about this, what the pros and cons are, says Fry. If you say that you prefer only buying black clothes from a specific retailer, then chat GPT will supposedly store that information in its memory the next time you ask for advice about what shirt to buy, giving you recommendations that align with your tastes. The reviews that ChatGPT features for products will pull from a blend of online sources, including editorial publishers like Wired, as well as user-generated forums like Reddit. Fry says that users can tell ChatGBT which types of reviews to prioritize

when curating a list of recommended products. One of the most pressing questions for online publishers with this new release is how likely affiliate revenue will work in this situation. Currently, if you read Wired's review of the best office chairs and decide to purchase one through our link, we get a cut of the revenue and it supports our journalism. How will affiliate revenue work inside of chat GPT shopping when the tool recommends an office chair that OpenAI knows is a good pick because Wired, among others, gave it good reviews? We are going to be experimenting with a whole bunch of different ways this can work, says Fry. He didn't share specific plans, saying that providing high-quality recommendations is Open AI's first priority right now and that the company might try different affiliate. revenue models in the future, end quote. So obviously a potential revenue driver for Open AI, but also let's underline that reveal that people are doing a billion web searches a week on

chat GPT already. Alarm bells at Google, no? Duolingo CEO Louis von On says the company will gradually stop using contractors to do work that AI can handle as part of an announcement to be an AI-first company. Quoting the verge. According to Von On, being AI-first means the company will need to rethink much of how we work and that making minor tweaks to systems designed for humans won't get us there. As part of the shift, the company will roll out a few constructive constraints, including the changes to how it works with contractors,

looking for AI use in hiring and in performance reviews, and that headcount will only be given if a team cannot automate more of their work. Von Onn says that Duolingo will remain a company that cares deeply about its employees and that this isn't about replacing duos with AI. Instead, he says that the changes are about removing bottlenecks so that employees can focus on creative work and real problems, not repetitive tasks. AI isn't just a productivity boost, Von On says. It helps us get closer to our mission to teach well. We need to create a massive amount of content, and doing that manually doesn't scale. One of the best decisions we made recently was replacing a slow man.

annual content creation process with one powered with AI. Without AI, it would take us decades to scale our content to more learners. We owe it to our learners to get them this content ASAP. Von Ones' email follows a similar memo Shopify CEO Toby Luckie sent to employees and recently shared online. In that memo, Luckie said that before Teams asks for more headcount or resources, they needed to show why they cannot get what they want done using AI, end quote. Amazon has launched 27 satellites for its Project Coiper Broadband Internet Program, the first batch of 3,236 satellites that it plans to send into low-Earth orbit. Quoting Reuters, sitting atop an Atlas 5 rocket from the Boeing and Lockheed Martin Joint Venture United Launch Alliance,

the batch of 27 satellites was lofted into space at 7 p.m. Eastern Daylight Time from the rocket company's launch pad at the Cape Canaveral Space Force Station. Bad weather scrubbed in initial launch attempt on April 9th. Coiper is arguably Amazon's biggest bet underway, pitting it against Starlink, as well as global telecommunications providers like AT&T and T-Mobile. The company has positioned the service as a boon to rural areas where connectivity is sparse or non-existent. The mission to deploy the first operational satellites has been delayed more than a year.

Amazon once hoped it could launch the inaugural batch in early 2024. The company faces a deadline set by the U.S. Federal Communications Commission to deploy half its constellation 1,618 satellites by mid-20206, but its slower start means Amazon is likely to seek an extension, analysts say. Hours or possibly days after the launch, Amazon is expected to publicly confirm initial contact with all of the satellites from its mission operations center in Redmond, Washington. If all goes this plan, the company said it expects to begin delivering service to customers later this year. ULA could launch up to five more co-oper missions this year. ULA, CEO Tori Bruno, told Reuters in an interview this month.

Amazon said in a 2020 FCC filing that it could begin service in some northern and southern regions at 578 satellites with coverage expanding toward Earth's equator as the company launches more satellites, end quote. Meta has launched a direct chat GPT competitor, and of course it has social stuff built in. The meta-AI app is a standalone app featuring a Discover feed where users can see AI interactions that friends have chosen to share. Quoting the verge, Meta stand-alone chat GPT competitor is mostly what you'd expect from an AI assistant. You can type or talk with it, generate images, and get real-time web results.

The biggest new idea in the meta-AI app is its Discover feed, which adds an AI twist to social media. Here you'll see a feed of interactions with meta-a-I that other people, including your friends on Instagram and Facebook, have opted to share on a prompt-by-prompt basis. You can like, comment on, share, or remix these shared AI posts into your own. The idea is to demystify AI and show people what they can do with it. Meta's VP of product Connor Hayes tells me. It may seem obvious that Meta is the first to add a social component to its AI assistant. It definitely won't be the last, though. Across the industry, AI chatbots and social media are converging. Elon Musk's X has already integrated closely with GROC. OpenAI, meanwhile, is planning to add a social feed to chat GPT. The meta AI app puts voice mode at the forefront,

an opt-in beta version makes Meta's AI voice more conversational like ChatGPT's advanced voice mode, though Meta's version currently lacks access to information from the web. So far, most people have experienced MetaI through its takeover of parts of Instagram, Facebook, and WhatsApp. Meta expects the majority of its usage to continue coming from places like the search bar on Instagram. Hayes says that the chatbot has reached almost one billion users this way, though he acknowledges that a standalone app is the most intuitive way to interact with an AI assistant, end quote.

and quoting CNBC. The event comes as Meta hosts its inaugural LamaCon developer event at its Menlo Park, California headquarters for its Lama family of AI models. Investors are on the lookout for signs that META's AI investments are having an immediate business impact. In January, the company announced plans to spend as much as $65 billion this year to expand its AI infrastructure, end quote. Yes, Lama Khan, Meta's first ever AI developer event. Zuck hosted Sacha Nadella on stage where he said that 20 to 30% of code in Microsoft's repositories was AI written, quoting TechCrunch. Nadella gave the figure after Zuckerberg asked roughly how much of Microsoft code is AI generated today. The Microsoft CEO said the company was seeing mixed results in AI generated code across different languages

with more progress in Python and less in C++. Microsoft's CTO Kevin Scott previously said he expects 95% of all code to be AI generated by 2030. When Nadella threw the question back at Zuckerberg, the meta CEO said he didn't know how much of meta's code is being generated by AI, end quote. Other headlines, meta debuted the Lama API for fine-tuning and evaluating the performance of its Lama models, available in limited preview with pricing yet to be announced. And this, quoting Wired, the end-to-end encrypted communications app WhatsApp, used by roughly 3 billion people around the world, will roll out cloud-based AI capabilities in the coming weeks that are designed to preserve WhatsApp's defining security and privacy guarantees while offering users access to message summarization and composition tools. Meta has been incorporating generative AI features across its services that are built on its open source large language model Lama, and WhatsApp

already incorporates a light blue circle that gives users access to the meta AI assistant. But many users have balked at this addition, given that interactions with the AI assistant and aren't shielded from meta the way end-to-end encrypted WhatsApp chats are. The new feature dubbed private processing is meant to address these concerns with what the company says is a carefully architected and purpose-built platform devoted to processing data for AI tasks without the information being accessible to meta, WhatsApp, or any other party. While initial reviews by researchers of the scheme's integrity have been positive, some note that the move towards AI features could ultimately put WhatsApp on a slippery slope.

End-to-end encrypted communications are only accessible. to the sender and receiver or the people in a group chat. The service provider, in this case, WhatsApp and its parent company meta, is boxed out by design and can't access users' messages or calls. This setup is incompatible with typical generative AI platforms that run large language models on cloud servers and need access to users' requests and data for processing. The goal of private processing is to create an alternate framework through which the privacy and security guarantees of end-to-end encrypted communication can be upheld while incorporating AI. Users have to opt in to using WhatsApp AI features, and they can also prevent people they're chatting with from using the AI features in shared communications by turning on a new WhatsApp control known as advanced chat privacy.

When the setting is on, you can block others from exporting chats, auto-downloading media to their phone and using messages for AI features, WhatsApp wrote in a blog post last week. Like disappearing messages, anyone in a chat can turn advanced chat privacy on and off, which is recorded for all to see. So participants just need to be mindful of any adjustments. Private processing is built with special hardware that isolates sensitive data in a trusted execution environment, a siloed, locked down region of a processor. The system is built to process and retain data for the minimum amount of time possible and is designed to grind to a halt and send alerts if it detects any tampering or adjustments, end quote. Have you heard the stories going around about the latest Open AI models being too, well, obsequious?

Like, basically, they kind of flatter you, kiss your butt. a bit. Well, OpenAI says GPT40's latest update is now 100% rolled back for free chat GPT users and is rolling back for paid users in a bid to fix the model's sycophancy. Quoting TechCrunch, we started rolling back the latest update to GPT40 last night. Sam Altman wrote in a post on X. It's now 100% rolled back for free chat GPT users and we'll update again when it's finished for paid users, hopefully later today. We're working on a additional fixes to model personality and will share more in the coming days. Over the weekend, users on social media blamed the updated model, which arrived toward the end of last week,

for making ChatGPT overly validating and agreeable. It quickly became a meme. Users posted screenshots of ChatGPT applauding all sorts of problematic, dangerous decisions and ideas. On Sunday, Altman acknowledged the problem and said that OpenAI would work on fixes ASAP and would share its learnings at some point, end quote. I keep flagging these stories because I feel like we might wake up someday soon and find ourselves already in a post-self-driving world. Waymo and Toyota have announced a preliminary partnership aiming to develop an autonomous vehicle platform and enhance Next Generation personally owned vehicles. Quoting CNBC, the companies will explore how to leverage Waymo's autonomous technology and Toyota's vehicle expertise to enhance Next Generation personally owned vehicles the two companies announced. The company said they aim to use the partnership to more quickly develop driver assistance and

autonomous vehicle technologies for personal vehicles. Toyota is the world's largest automaker by sales. Waymo co-CEO co-CEO Tecendra Mawakana said the strategic partnership could also result in the Google-owned company incorporating Toyota's vehicles into our ride-hailing fleet. The Toyota tie-up is the latest automotive partnership for Waymo. The self-driving company has previously worked with automakers such as Jaguar Land Rover, Stalantis predecessor Fiat Chrysler, Daimler, Dymler, Hyundai Motor, and China's Geyli. The partnerships, many of which touted long-term tie-ups,

largely resulted in automakers producing modified vehicles for testing or for Waymo to use in its fleets. The partnership with Toyota will not affect Waymo's plans to deploy Hyundai and Geely vehicles through the Waymo One ride hailing service in the future. A spokesman for the alphabet-owned company told CNBC, Waymo is now serving 250,000 paid rides per week up from 200,000 in February before Waymo opened in Austin and expanded in the San Francisco Bay Area in March. Waymo is already running its commercial driverless ride hailing services in the San Francisco, Los Angeles, Phoenix, and Austin regions, end quote. Have you noticed the quality of Google Play apps getting better?

According to app figures, Google Play now hosts around 1.8 million Android apps down 47% from the 3.4 million. at the start of 2024. This is likely due to raising minimum quality requirements back in July of last year. Quoting TechCrunch. From the start of 2024 to the present, the Android app marketplace went from hosting about 3.4 million apps worldwide to just around 1.8 million, according to a new analysis by app intelligence provider app figures. That's a decline of about 47 percent, representing a significant purge of the apps that have been available to Android users globally. The decline is not part of some larger global trend, the firm also notes. During the same period, Apple's iOS App Store went from hosting 1.6 million apps to now just around

1.64 million apps, for instance, a slight increase. In Google's case, the decline in apps could be a relief for Android device owners who have had to sort through scummy, spammy, and otherwise poor-quality apps to find the best ones to install. The reduction could also help developers who have had to fight for visibility. Over the years, Google plays less stringent requirements for app review have led to the marketplace being overrun with lower quality apps. While Apple continues to enforce strict app review measures before publication, Google often relies on automated checks combined with malware scans to speed up the app review process. It tends to have a shorter app review period as a result of its lighter touch in terms of human

review. In July 2024, though, Google announced it would raise the minimum quality requirements for apps which may have impacted the number of available Play Store app listings. Instead of only banning broken apps that crashed, wouldn't install or run properly, the company said it would begin banning apps that demonstrated, quote, limited functionality and content. That included static apps without app-specific features such as text-only apps or PDF file apps. It also included apps that provided little content like those that only offered a single wallpaper. Additionally, Google banned apps that were designed to do nothing or have no function, which may have been tests or other abandoned developer efforts.

Reached for comment, Google confirmed that its new policies were factors here, which also included an expanded set of verification requirements, required app testing for new personal developer accounts, and expanded human reviews to check for apps that try to deceive or defraud users. In addition, the company pointed to other 2024 investments in AI for threat detection, stronger privacy policies, improved developer tools, and more. As a result, Google prevented 2.36 million policy-violating apps from being published on its Play Store and banned more than 158,000 developer accounts

that had attempted to publish harmful apps, it said. One factor Google didn't cite was the new traitor status rule enforced by the EU as of this February, which began requiring developers to share their names and addresses in the apps listing. Those who failed to do so would see their apps removed from EU app stores. It's worth pointing out that Apple also began requiring trader status information in February and did not see a decline in available apps as a result, end quote. I do want to note that Amazon walked back that rumor of showing tariffs in shopping results. Unclear what happened here, but CNN had sources that said, President Trump called Jeff Bezos on Tuesday morning to complain about reports that Amazon was adding tariff prices to listings.

For its part, Amazon says Amazon Hall, quote, considered the idea of listing import charges on certain products, but, quote, this was never a consideration for the main Amazon site. So was it, as we discussed, not wanting to catch the ire of the administration, which seemed to happen immediately, I don't know. On the rest of the world, possibly decoupling from Silicon Valley front on the whole sovereign tech stack front, Politico says that the EU's draft international digital strategy set for a release on June 4th acknowledges that decoupling from U.S. tech is, quote, unrealistic, and instead calls for strategic. Alliances. Quote, tech competitiveness is an economic and security imperative for all aspiring to durable wealth and stability, says a draft version dated April 9th. Yet when it comes to

dominant players such as the U.S., quote, decoupling is unrealistic and cooperation will remain significant across the technological value chain, the draft says. It cites China, as well as Japan, South Korea, and India as countries with which collaboration will also be essential. The pitch for strategic tech alliances with like-minded countries to team up on research and generate greater business opportunities for the bloc's companies comes in stark contrast to growing calls for a move toward protectionism. For Europe, quote, business as usual is no option, wrote Margett Chucky, a former Dutch liberal member of the European Parliament, who is a leading voice on tech called on the block to, quote, end its debilitating dependence on American tech groups and take concrete steps

to shield itself from the growing dangers of this new tech-fueled geopolitical landscape, end quote. In Brussels, the idea of a Eurostack, an ambitious industrial plan to break free from U.S. tech dominance is gaining steam with key lawmakers throwing their weight behind the proposal. The draft strategy backs international engagement on critical technologies such as quantum and chips, as quote, the growing complexity of semiconductor supply chains and geopolitical uncertainty necessitate a tailored, country-specific approach. The EU has been scrambling to fix, among other things, a risky reliance on China for low-tech chips, end quote. But remember, one of the key things I worry about here with a sovereign tech movement is that a big tech American platform, say, could no longer

assume a singular global market. It would have to consider multiple sort of walled-off markets. And in the worst case scenario, that might even mean losing access to markets. I'm thinking of someone like, say, Microsoft, losing access to, say, the European market. Well, Microsoft President Brad Smith says his company would take the U.S. government to court if necessary to protect European customers' access to its services. Quoting the FT. Brad Smith, Microsoft's president, said European leaders were shocked when the Trump administration temporarily suspended military and intelligence support to Ukraine. The cloud computing and software giant on Wednesday responded with new commitments to European governments about, quote, continuity of access. Quote, we as a company need to be

a source of digital stability, said Smith, who is also Microsoft's vice chair and top lawyer. Microsoft's new pledge includes five digital commitments to Europe. The big tech group said it would contest any government order to cease cloud services to European customers, including through the courts. It also promised to have its cloud computing service in the continent overseen by a European board of directors and operating under European law. The company is planning to boost its European cloud and artificial intelligence operations by increasing its data center capacity in Europe by 40% over the next two years, expanding operations in 16 countries. It expected to spend, quote, tens of billions of dollars a year on European data centers, Smith said, amid recent

speculation that the company was pulling back on some of these investments. The Seattle-based company is the first large American tech business to proactively try to reassure European customers amid escalating trade tensions and calls for more European tech sovereignty, including demands to exclude American companies from public contracts. President Trump has fueled anxiety among European governments and companies over privacy and data access and even prompted concerns that the U.S. could suspend or block the operations of American tech companies in Europe. Smith said a suspension was unlikely, quote, there is a strong consensus in Washington that wants to see American digital technology flow to Europe, end quote. But he acknowledged the topic

had been on the mind of European leaders, especially after Trump's temporary suspension of military and intelligence support to Kiev. They asked themselves about how they're getting their defense and security protection more broadly, Smith said, I think it's therefore important for us to make clear that Europe can count on us, end quote. Whoa, a U.S. judge has ruled that Apple violated a 2021 court order to open the app store to third-party payment options and has referred the case for a criminal investigation. You heard that right. Quoting Bloomberg, U.S. District Judge Yvonne Gonzalez-Rogers cited Wednesday with Fortnite developer Epic Games over its allegation that the iPhone maker failed to comply with an order she issued in 2021 after finding

the company engaged in anti-competitive conduct in violation of California law. Gonzalez-Rogers also referred the case to federal prosecutors to investigate whether Apple committed criminal contempt of court for flouting her 2021 ruling. The U.S. Attorney's Office in San Francisco declined to comment. The changes the company must now make could put a sizable dent in the double-digit billions of dollars in revenue the App Store generates each year. Apple is potentially facing another multi-billion dollar hit from losing payments Google makes to be the default search engine for its Safari browser, which is the subject of an ongoing Justice Department antitrust case against

the alphabet unit. After several weeks of hearings last year and this, Gonzales Rogers concluded Wednesday that Apple, quote, willfully violated her injunction, quote, it did so with the express intent to create new anti-competitive barriers, which would, by design, and in effect, maintain a valued revenue stream. A revenue stream previously found to be anti-competitive, she wrote in her 80-page ruling, that it thought this court would tolerate such insubordination was a gross miscalculation, end quote. Epic Games chief executive officer Tim Sweeney called the ruling a huge victory for developers, saying in a phone call with journalists, it forces Apple to compete with other payment services rather than blocking them, end quote.

Actually, quoting Tim Sweeney on X, We will return Fortnite to the U.S. iOS App Store next week. And Epic puts forth a peace proposal. If Apple extends the court's friction-free Apple tax-free framework worldwide, we'll return Fortnite to the App Store worldwide and drop current and future litigation on the topic, end quote. We'll see if anything comes of that. Apple, of course, says it disagrees with the court's injunction and plans to appeal, but will comply with the order in the meantime. Meanwhile, CNBC, was reporting that Apple's Phil Schiller wanted Apple to comply with the court order, but Tim Cook ignored him. And this, quote, Apple vice president of finance Alex Roman, quote, outright lied to the court

about when Apple had decided to levy a 27% fee on some purchases linked to its app store. Neither Apple nor its counsel corrected the now obvious lies, Judge Gonzalez-Rogers wrote, saying that she considers Apple to, quote, have adopted the lies and misrepresentations to this court. The decision is a striking repudiation of Apple's conduct in the Epic Games trial, which was decided in 2021 and appealed in 2023. While Apple won the vast majority of counts in the original trial, Epic Games did win some concessions tucked inside a 180-page order. Rogers originally ordered the company to make changes to its app store, allowing software developers to link to their websites inside of iPhone apps for customers to make purchases outside of Apple's ecosystem. On Wednesday, Rogers accused Apple of willfully trying to violate her ruling, and she held the company in contempt. Rogers wrote that it was expected under her ruling that those kinds of off-app purchases would not have an Apple commission.

But Apple introduced new policies in 2024 that collected a 27% commission from some of those purchases, only a slight discount from the 30% Apple usually collects from in-app purchases. Rogers said nearly every Apple decision on its app linking policies was anti-competitive. Rogers wrote that Apple presented evidence to the Court of Internal Deliberations about its rule that were, quote, Taylor made for litigation instead of the company's actual internal discussions. Quote, in stark contrast to Apple's initial in-court testimony, contemporaneous business documents revealed that Apple knew exactly what it was doing and at every turn chose the most anti-competitive option, Rogers wrote.

To hide the truth, Vice President of Finance, Alex Roman, outright lied under oath, end quote. Rogers also accuses Apple of withholding documentation of a June 2023 meeting, including CEO Tim Cook, about how they would comply with the 2021 court order. Rogers said that Apple hid the existence of the meeting from the court until 2025. She also said that Apple abused privilege in order not to share documents that it was supposed to. Apple had a, quote, desire to conceal Apple's real decision-making process, particularly where those decisions involved senior Apple executives, Rogers wrote. Former Apple senior vice president and current fellow Phil Schiller did not want Apple to take a commission on web links, but

Cook ignored him, Rogers said. Cook chose poorly, Rogers wrote. The judge ordered effective immediately for Apple to stop imposing its commissions on purchases made for iPhone apps through web links inside an app. She also ordered Apple to pay Epic Games's attorney's fees over this specific issue. This is an injunction, not a negotiation. There are no do-overs once a party willfully disregards a court order, Rogers wrote, end quote. Meanwhile, you might have heard about this, quoting the journal. About a month ago, with Tesla's stock sinking, and some investors irritated about Elon Musk's White House focus, Tesla's board got serious about looking for Musk's successor. Board members reached out to several executive search firms to work on a formal process for finding

Tesla's next chief executive, according to people familiar with the discussions. Tensions had been mounting at the company. Sales and profits were deteriorating rapidly. Musk was spending much of his time in Washington. Around that time, Tesla's board met with Musk for an update. Board members told him he needed to spend more time on Tesla, according to people familiar with the meeting, and he needed to say so publicly. Musk didn't push back. Tesla has been on a losing streak in the month since Musk, its visionary chief executive began spending much of his time helping President Trump's slash federal spending. Last week after the company said its first quarter profit had plunged 71 percent, Musk told investors he would soon pivot back to his job at Tesla.

Starting next month, he said on a conference call about earnings, I'll be allocating far more of my time to Tesla. The board narrowed its focus to a major search firm, according to people familiar with the discussions. The current status of the succession planning couldn't be determined. It is also unclear if Musk, himself a Tesla board member, was aware of the effort, or if his pledge to spend more time at Tesla has affected succession planning. Musk didn't respond to requests for comment. Tesla didn't provide a statement before publication. Hours after this article was published, Tesla issued a denial on X.

Musk also criticized the article in a post on X. The CEO of Tesla is Elon Musk and the board is highly confident in his ability to continue executing on the exciting growth plan ahead. Tesla chair, Robin Denham, said in a statement posted on X. Early last year, after some two decades of running Tesla, Musk confided to someone close to him in late-night techs, that he was frustrated to still be working nonstop at the company, especially after a Delaware judge had struck down his multi-billion dollar pay package. Last spring, he told that person that he no longer wanted to be CEO of Tesla, but that he was worried that no one could replace him atop the company and sell the vision that Tesla isn't just an automaker, but the future of robotics and automation as well.

Musk has complained both in public and private that despite owning roughly 13% of the company, he has been working without pay for the last seven years. The Tesla board recently formed a special compensation committee to address CEO compensation. Some Tesla employees said that the first time they had heard from Musk in months was at an all-hands meeting in March, streamed to all ex-users, where he tried to reassure employees and persuade them not to sell their shares. If you read the news, it feels like Armageddon, he told them, I can't walk past the TV without seeing a Tesla on fire, he said, referring to vandalism at Tesla showrooms and charging stations. There are times when there are rocky moments, a little bit of

stormy weather, but I'm here to tell you that the future is bright and exciting. What I'm saying is, hang on to your stock, end quote. Meta reported Q1 revenue up 16% year on year and family daily active people up 6% to 3.4 billion on average for March of 2025, yada, yada. More interesting was that Mark Zuckerberg said that WhatsApp now has more than 3 billion monthly active users up from 2 billion in February 2020 and 1.5 billion in January 2018. Also, meta increased its 2025 CAPX to a range of $64 to $72 billion due in part to higher infrastructure costs up from its prior outlook of $60 to $65 billion. Meta's stock jumped 5% on the news, so I guess Wall Street is cool with all of that spending

continuing. Microsoft Q3 revenue was up 13% year-on-year to $71.1 billion versus $68.42 billion estimated. Microsoft 365 commercial products and cloud services up 11% and Q4 revenue forecasts were above estimates. The all-important Azure and other cloud services revenue was up 33% above estimates. And Microsoft said, GitHub co-pilot has more than 15 million users up 4x year-on-year. GitHub CEO Thomas Domke said its code review agent launched in April has handled more than 8 million pool requests. This morning, Microsoft raised the prices of its Xbox Series S and X consoles, Xbox controllers, and some new Xbox games globally.

The Xbox Series S is up $80 to $380, quoting the verge. The Xbox Series X is getting bumped by $100 to $599. dollars. Microsoft is also planning to adjust the pricing of some of its new first-party Xbox games this holiday season up to $799. Xbox wireless controllers and headsets are also having their recommended retail pricing adjusted in the U.S., with the base controller moving to $649.9 and the Xbox wireless headset now at $119.99. You can already find both of these accessories at this type of pricing right now. Microsoft last hiked the price of its Xbox Series X console in June in 2020 as part of price increases that also impacted Xbox Game Pass.

The subscription service also saw a price increase last year alongside the launch of a new standard tier. The price rises come amid uncertainty over the Trump tariffs and just a day after Microsoft's CEO Sachin Adela revealed that Microsoft, quote, ended the quarter as the top publisher by pre-orders and pre-installs on both Xbox and PlayStation Store during the company's Q3 fiscal earnings call. Microsoft also saw PC GamePass revenue increase by 45% year over year, end quote. And meta updated the Ray Ban meta Smart Glass's privacy policy on April 29th to enable the

Hey Meta voice command by default. Cool, right? Downside is they will also be storing voice recordings for a year. So you know how you always have that thought in the back of your mind that maybe somehow Meta is listening in on what you're saying to target you with ads? Now that might actually be true, question mark? Quoting the verge. Meta is making a few notable adjustments to the privacy policy for its Rayband Meta smart glasses in an email sent out on April 29th to owners of the glasses the company outlined two key changes. First, meta AI with camera use is always enabled on your glasses unless you turn off Hey Meta, the email said. The latter refers to the hands-free voice command functionality. That said, spokesperson Albert Iden tells the

verge the photos and videos captured on Rayban meta are on your phone's camera role and not used by meta for training, including photos or videos captured by using the Hey Meta, take a photo slash video voice command. If you share those photos to a product, for example, meta AI, cloud services, or a third-party product, then the policies of that product will apply. Second, meta is taking after Amazon by no longer allowing RayBan meta owners to opt out of having their voice recordings stored in the cloud. The option to disable voice recording storage is no longer available, but you can delete recordings any time in settings, the company wrote. In its voice privacy notice, Meta states that voice transcripts and stored audio recordings

are otherwise stored for up to one year to help improve Meta's products. If the company detects that a voice interaction was accidental, those recordings are deleted after a shorter 90-day window. The motivation behind these changes is clear. Meta wants to continue providing its AI models with heaps of data on which to train and improve subsequent results. Some users began noticing these policy changes in March, but, at least in the United States, Meta says they went into effect as of April 29th. Earlier this month,

the company rolled out a live translation feature to the Rayban meta product, and just yesterday, meta rolled out a standalone meta-AI app on smartphones to more directly compete with OpenAIs chatGPT, Google Gemini, Anthropics Claude, and other AI chatbots. The company is reportedly planning a higher-end pair of Rayban meta-glasses for release later in 2025, end quote. Finally today, from the verge, this headline kind of says it all. Would you wear a smart ring that proves you're not a cheater? Quote, imagine this. You wake up, roll over, take your smart ring off its charger, and slip it onto your finger.

After eating breakfast and getting ready for work, you kiss your lover goodbye and head out. Everything's hunky-dory until you look down at your finger. The LED light on your smart ring is flashing pink. It means something's up with your lover, probably with an elevated heart rate. You pull out your phone, open an app, and read detailed AI-generated insights that indicate your lover is aroused. You furrow your brow because you're not at home. Something is up. What I've written is, as of right now, science fiction.

However, if the raw ring sees the light of day, this could become a real product you can buy, at least in theory. You may have seen some coverage of the raw ring in the past few weeks. The smart ring has been described as a dystopian loyalty tracker that can help you catch a cheating partner in the act. It's not a product that actually exists yet, but the idea was developed by the folks behind Raw, a dating app that aims to cut down on catfishing and ghosting by making its users upload unfiltered real-time dual camera selfies. The Raw Ring, CEO Marina Anderson says, is meant to be an extension of that app. For what it's worth, App Store reviews say the dating app is also a work in progress. The idea behind the ring was to give couples more ways to explore each other's feelings on a deeper level and build more trust, says Anderson, noting that Raw chose.

a ring because it's been a cultural symbol of trust for thousands of years. Almost 50% of marriages now end in divorce. We would like to build the ring as a symbol of trust again. You can't buy it yet, and it's unclear if and when you can, but Anderson has a blueprint of how it'll work. The smart ring will have an optical sensor to measure heart rate and heart rate variability, a skin temperature sensor, plus an accelerometer and gyroscope for detecting movement. It will also have a noise-canceling memm's microphone to analyze voice tone. The ring will utilize a dual-level AI, one that's on device to handle real-time processing of voice and biometric data, and another in the cloud to do, quote, deeper trend analysis. Anderson claims that AI will be able to differentiate

between vigorous exercise, arousal, and just regular activity by analyzing multiple data points. Physiological stress, for example, is signaled through metrics like heart rate and temperature, while emotional anxiety is indicated by shifts in tone or irregular movement. Lastly, Anderson claims there will be a customizable LED light and Chi wireless charging. On paper, this is all feasible. Aspects of the raw ring are in products you can buy now. Smart rings like the ORA ring and many of its new competitors track stress and can automatically detect exercise from other types of movement,

though smart rings aren't always the best at differentiating between sexual activity, intentional exercise and you huffing and puffing up the stairs. B, an AI wearable recorder and the now-defunct Amazon Halo, also used AI to analyze vocal tones. There are gadgets like the Bond Touch bracelet, which allow long-distance partners to send each other vibrations to show their thinking of each other. Raw Ring is simply proposing to mash together these disparate parts into a single device. But the question with the Raw Ring isn't whether it's possible to make, it's whether we should, end quote. Rockstar Games has delayed the release of Grand Theft Auto 6 from what was expected to be this fall to May 26, 2026, saying it needs extra time to deliver the expected quality. I said all year that this launch would be the cultural event of the year by far, and now, quoting the verge.

In a message posted on the Rockstar Games website, the development team apologized for the delay. We are very sorry that this is later than you expected. The interest and excitement surrounding a new Grand Theft Auto has been truly humbling. for our entire team. We want to thank you for your support and your patience as we work to finish the game. With every game we have released, the goal has always been to try and exceed your expectations and Grand Theft Auto 6 is no exception. We hope you understand that we need this extra time to deliver at the level of quality you expect and deserve. We look forward to sharing more information with you soon, end quote. And quoting Bloomberg. Rockstar Games, the studio behind the hit, has revealed little about

GTA 6 aside from a teaser trailer for the game in late 2023. The trailer, which has topped 200 million views on YouTube, shows the game is set in a fictional version of Miami. The game, which has already missed multiple deadlines, was expected to be the biggest release of 2025 and one of the most lucrative games ever, end quote. And quoting Reuters. The previous entry, Grand Theft Auto 5, released in 2013, has sold more than 200 million copies, making it one of the best-selling video games of all time. The delay. now pushes the game out of Take 2's fiscal 2026 release window and into 2027, which will likely bring down booking expectations for the next fiscal year. GTA6 was also expected to be a major

driver of overall video game industry growth this year after the market underwent a downturn following pandemic highs, end quote. Revenue round up to the tune of Mystery Science Theater's robot roll call. Apple reported Q2 revenue up 5% year-on-year-on-year, but China. China sales were down 2% year-on-year. On its conference call, Apple said that a majority of iPhones sold in the U.S. in Q3 will come from India, while nearly all of its other devices will come from Vietnam. They also planned to source more than 19 billion chips from the U.S. this year, including from TSM's expanding Arizona facilities, as it seeks to lessen its reliance on China.

They also reported Q2 revenue from services, which includes the App Store, Apple TV Plus, and Apple music up 12% year-on-year to 26.65 billion, a new record, but below the 26.7 billion estimated. And Tim Cook said that the company estimates tariffs will add $900 million in costs in its Q3, assuming current tariff rates and policies do not change. Amazon reported Q1 revenue up 9% year-on-year, operating income up 20% year-on-year to $18.4 billion, but, crucially, forecast Q2 operating. income below estimates. Quoting the times, Amazon told investors to expect sales of 159 billion to 164 billion and for operating profits to shrink to as low as 13 billion. Amazon added tariff and

trade policies to the list of factors it says can make its forecasts uncertain. The results were mixed compared with Wall Street's expectations. Amazon's stock price was down more than 3% in aftermarket trading following the earnings release. Obviously, none of us know exactly where tariffs will settle or when. Andy Jassy, the chief executive of Amazon said on a call with investors. He said the company is pretty maniacally focused on keeping prices down by purchasing extra inventory in advance of tariffs and will be helping sellers on Amazon's marketplace do the same, end quote. Microsoft says all new accounts will be passwordless by default, asking people to use more secure methods like pass keys, push notifications, and security keys. Quoting the verge, the new note. The new notice. The new,

password initiative by Microsoft is accompanied by its recently launched optimized sign-in window design with reordered steps that flow better for a passwordless and pass-key-first experience. Although current accounts won't have to shed their passwords, new ones will try and leave them behind by not prompting you to create a password at all. As part of this simplified U.X, we're changing the default behavior for new accounts. Brand-new Microsoft accounts will now be passwordless by default. new users will have several passwordless options for signing into their account and they'll never need to enroll a password. Existing users can visit their account settings to delete their password.

With today's changes, Microsoft is renaming World Password Day to World Pass Key Day instead and pledging to continue to work implementing pass keys over the coming year. This time last year, the company implemented pass keys into consumer accounts. Microsoft says it's seeing nearly a million pass keys registered every day. and that past key users have a 98% success rate of signing in versus 32% for password-based accounts, end quote. Tamu appears to have removed all products in the U.S. version of its online store that ship directly from China to U.S. consumers, confusing suppliers and customers alike, quoting Wired. The change seemingly happened earlier this week, just days before a trade loophole that allows American consumers to buy products from China without paying tariffs is set to disappear

as part of sweeping new import duties imposed by U.S. President Donald Trump. Over the past week, Temu has rolled out a number of whirlwind changes to its platform as it grappled with the impacts of Trump's trade war on its U.S. business, which is built over the past three years by offering products at astonishingly low prices and promoting itself as a haven for bargain lovers. First, Temu announced that it would begin raising prices on products shipped from China, starting on April 25th, but then things got a lot more confusing. Soon afterward, Temu began displaying a separate import charge on orders from U.S. customers, seemingly as part of an effort to emphasize the financial impacts of the tariffs. Other retailers have also

adopted the tactic, but it has drawn harsh criticism from Trump. By Tuesday, U.S. shoppers realized that Temu had apparently decided to simply block U.S. users from seeing any product listings for items currently located in China or anywhere else outside the U.S. The version of Temu's website and app for the United States now appears entirely filled with products marked with a local label, meaning they are exempt from tariffs because they were shipped into the country before the new import duties went into effect. Prior to this week, consumers had the ability to choose between products with the local label and those without it, the latter of which are typically shipped from China via air cargo after a purchase is made. Things are in chaos right now.

Ever since the tariffs kept changing, our business has been heavily affected, says a Temu seller in China who specializes in furniture and home decor and asks to remain anonymous for privacy reasons. Temu has recently transitioned its U.S. operations to a local fulfillment model. This means that all sales in the U.S. are now handled by locally based sellers with orders fulfilled from within the country. An emailed statement from Temu said, confirming that the platform is cutting its ship from China strategy. Despite the operational shift, Temu's pricing for U.S. consumers remains unchanged, the statement also said. The shift has brought Temu closer in line with what consumers can expect from competitors like Amazon, says Josas Kasekouinus.

an independent e-commerce industry analyst. Today, Temu looks a lot like Amazon, because everything you buy on Temu today will come to you from their warehouse in the U.S. and probably in just a few days, he says. It also means that many U.S. shoppers are suddenly faced with a much narrower selection of goods on Temu, and they are not happy about it. On social media sites like Reddit, users have reported seeing hundreds of products they saved on wish lists and in their shopping carts suddenly become sold out overnight, end quote. From the waking up to a self-driving world file, Aurora has launched its commercial driverless trucking service, making deliveries between Dallas and Houston, and with plans to expand to El Paso and Phoenix this year. Now, self-driving trucks on highways was something I felt would become more of a reality long before this as the problem to solve, I thought was easier, right?

But nonetheless, quoting the verge. After years of testing and validation, Aurora says its first fully autonomous track tractor trailers are operating on public highways in Texas. The company's class eight trucks are now making customer deliveries between Dallas and Houston, having already completed 1,200 miles without a driver, Aurora said. The clients for these initial trips are Uber freight, the ride-hailing company's trucking brokerage, and Hirschbach Motor Lines, a carrier that delivers time and temperature-sensitive freight. Aurora's CEO Chris Irmson said he rode in the backseat during the first truck's inaugural ride, which he called the honor of a lifetime. We founded Aurora to deliver. We founded Aurora to deliver the benefits of self-driving technology safely, quickly, and broadly,

Ermson said in a statement, now we are the first company to successfully and safely operate a commercial driverless trucking service on public roads. Aurora said it plans to expand its driverless service to El Paso and Phoenix by the end of 2025. Driverless trucks were once expected to precede robotaxies and personally owned autonomous vehicles in mass adoption, considering that highways are vastly less complex than city and residential streets. But self-driving truck operators have run into hurdles involving the technology and regulations

that have delayed their public debut. Some companies like Embark trucks too simple and locomation have gone out of business, while others have cut plans to deploy driverless trucks as timelines have stretched into the future, and funding has dried up. Moreover, public opinion toward autonomous vehicles has trended downward thanks in part to missteps from companies like Tesla and Cruise. But like Waymo, Aurora has placed its hopes on a measured conservative approach to commercialization, as well as an emphasis on safety. Founded in 2017 by alumni of Uber, Tesla and Waymo, Aurora had planned to deploy its fully autonomous trucks in 2024, but those plans got delayed until this year, with the company continuing to tweak its autonomous system for

surface street driving and construction sites. Aurora says its technology presents a possible solution to the challenges currently facing the trucking industry, such as a trucker shortage, high turnover rates, and increasingly expensive operating costs. The company says its system can address these specific problems while also reducing labor costs and heightening safety on the highway. Aurora has spent four years conducting supervised pilot halls, mostly in Texas, where it delivered over 10,000 customer loads across 3 million autonomous miles. The company says it has also demonstrated capabilities such as predicting red light runners, avoiding collisions, and detecting pedestrians in the dark hundreds of meters away, and it has forged partnerships with a bunch of leading

players in the trucking industry, including Continental, Volvo, Uber, and others. The need to start charging customers for deliveries is evident if you look at Aurora's earnings. In its most recent report, the company reported a net loss of $748 million for 2024 down from $796 million the previous year. While the loss decreased, Aurora's revenue estimates have declined. Aurora expects to report its first quarter earnings on May 8th, end quote. Time for the weekend long-rate suggestions. And first up, what happens when AI upends an entire field? Where do the experts go? What do young people do? Researchers studying natural language processing were among the very first to find out. This oral history from Quanta looks at how LLMs and

Chatt GPT completely changed the field over the last five or so years. Quote, this is Christopher Callison Birch. It's an oral history, remember? I got early access to the GPT3 beta and was actually playing with it myself. I'm trying out all these things that my recent PhD students had done as their dissertations and just realizing, oh my God, the thing that had taken a student five, years. Seems like I could reproduce that in a month. All these classical NLP tasks, many of which I had touched on or actively researched throughout my career, just felt like they worked in one shot, like done. And that was just really, really shocking. I sometimes describe it as having this career existential crisis. This is Vecusi Maravate, quote, I don't know how many tutorials I gave on

LLMs in 2023. On one hand, you've been trying to talk to people for years and say, there's interesting stuff that's happening here. Then all of the sudden, it's just a complete waterfall of come explain this to us. Sam Bowman. It goes from a relatively sleepy field to suddenly, I'm having lunch with people who were meeting with the Pope and the president in the same month. Emily M. Bender. Between January and June, I counted five workdays with no media contact. It was non-stop. Is Beltegay? There is a point where you realize that in order to continue advancing the field, you need to build these huge expensive artifacts like the large Hadron Collider. You can't advance experimental physics without something like this.

I was lucky to be at AI2, which generally has more resources than most academic labs. ChatGPT made it clear that there's a huge gap between OpenAI and everybody else. So right after, we started thinking about ways we can build these things from scratch. And this is exactly what happened, end quote. And then from Bloomberg, the life of the most used city bike in New York, City. City bike, that's our ride-sharing bike service here in the city. Quote, City bike number 3-2606 was born on October 15th, 2017 at 1108 a.m. Its first journey began at 3rd Street and Prospect Park West in Park Slope, Brooklyn,

on a glorious 70-degree mid-October day. The ride ended at Grand Army Plaza 28 minutes later, barely a half mile away. The rider was not a city bike member, so I suspect a tourist used the weather as an excuse to take a leisurely ride around Prospect Park. The leaves were probably just starting to change. Over its 806 days of service, about two years and two months, City Bike number 3-2606 was ridden 8,624 times, or an average of 10 times per day. This makes City Bike 3-2606 the most used traditional bike we know of in the public data set as of 2020. In 2020, CityBike stopped publishing the unique bike identifier for each trip taken. E-bikes are excluded from this data. As the crow flies, the total distance covered was more than

7,000 miles, but this is an underestimate of how far City Bike 3-2606 actually traveled, because the data only tells us where the bikes docked, not their on-street routes. Still, that 7,000 miles would be enough to go from Times Square to the Santa Monica Pier in Los Angeles and back again, plus a final jaunt up to Burlington, Vermont. City Bike 3-2606 did that distance just on New York City's streets, and every single one of those 7,000-plus miles was powered by human legs. We know some other things about City Bike 3-2606 and its humans from the data. The median age of its rider was 37 years old. The youngest, or at least the youngest rider, to legally unlock the bike from a dock,

was a series of 16-year-olds on four separate occasions, and the oldest was 85 years young. After 8,623 trips, the bike's final recorded ride was on December 28, 2019, a short hop through the Lower East Side for three blocks. It took a hair over two minutes. It would have been a six-minute walk, but 32606's last act was still a noble one. It saved someone a couple of minutes, end quote.