Tech Brew Ride Home - Fri. 10/11 - Cybercab

Episode Date: October 11, 2024

All the headlines from Tesla’s robotaxi event. AMD unveils the chip it wants to use to go toe to toe with Nvidia. Are people actually making money building GPTs or not so much? And in the longreads, rebuilding Notre Dame as period accurately as possible. Sponsors: Go to Hensonshaving.com/ride and enter RIDE at checkout to get 100 free blades with your purchase. (Note: you must add both the 100-blade pack and the razor for the discount to apply.) Links: Tesla reveals 20 Cybercabs at We, Robot event, says you’ll be able to buy one for less than $30,000 (TechCrunch) Tesla's Cybercab Is Here (Wired) Tesla Robovan has something the Cybertruck definitely does not (T3) AMD launches AI chip to rival Nvidia’s Blackwell (CNBC) Xbox will sell games directly in the Android app next month (The Verge) OpenAI’s GPT Store Has Left Some Developers in the Lurch (Wired) Weekend Longreads Suggestions: Google’s Nobel Haul Puts Spotlight on Changing Focus of Scientific Research (Bloomberg) The Miraculous Resurrection of Notre-Dame (GQ) 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. Welcome to the Tech meme right home for Friday, October 11th, 2024. I'm Brian McCullough today. All the headlines from Tesla's Robotaxy Event. AMD unveils the chip it wants to use to go toe-to-to-to-with Nvidia. Are people actually making money building GPTs or not so much? And in the long reads,

rebuilding Notre Dame as period accurately as possible. Here's what you miss today in the world of tech. So the Robotaxy event was late last night. And when I say late, I mean late. It was almost an hour late starting. Elon Musk revealed the Tesla cybercab saying you will be able to buy one for under $30,000 and that he expects the Robotaxie to be in production before 2027. Quoting TechCrunch, Tesla has finally revealed its cybercab and it looks like a smaller, sleeker, two-seater cyber truck. And while many were expecting there to be at least one prototype of a robot taxi with no steering wheel or pedals, Tesla CEO Elon Musk delighted his fans with a lineup of 20 vehicles. The flashy Wii robot event took place at Warner Brothers Discovery Studio on Thursday.

Before walking on stage, Musk walked over to a robotaxie, which opened its gallwing doors, and did a short demo around the well-maintained streets of the Hollywood studio. Musk repeated previous claims that the cost of autonomous transport will be so low. It will be akin to, quote, individualize mass transit. He said he believed the average operating cost of the cyber cab will be over time around 20 cents per mile. and you will be able to buy one, Musk said, adding that the cost of the vehicle would be below $30,000. Musk also noted he expects Tesla to start doing, quote, unsupervised FSD in Texas and California next year, with the Model 3 and Model Y. He acknowledged that he's too optimistic about timelines sometimes,

but said he expects the cybercab to be in production by 2026 or before 2027. Fans cheered when Musk said they would be able to test out the cybercabs themselves at the event. They have like 20 of them driving around the entire lot totally unsupervised. One event goer told TechCrunch, the Robotaxi also doesn't have a plug-in charger and instead has inductive charging, which is a sort of wireless charging according to Musk, end quote. And quoting wired. Video renderings showed robots cleaning out the interior of a cyber cab pointing to a solution to the off-sighted autonomous taxi problem, how to keep the things clean without the assistance of a human driver.

The Robotaxie would also be charged wirelessly through inductive. charging, Musk said, but a timeline for both tech features went unmentioned. The event concluded with Optimus, a humanoid robot that Musk has said could eventually make the automaker some $25 trillion by becoming, quote, the biggest product ever of any kind. The robot was making progress, Musk said, to prove it five bots illuminated by lights danced in a nearby gazebo. More Optimus bots mingled with the crowd after the presentation serving drinks at the bar and posing for photos, end quote. And there was one more thing. Back to TechCrunch. Tesla unveiled a surprise robo van at the event, a sleek-looking autonomous bus that can carry up to 20 people and also transport goods.

No timelines were revealed for this vehicle, only a hand-waving of a future that could, quote, change the look of the roads, end quote. And quoting from T3. Where the cyber truck is all sharp edges and with a seemingly unfinished aesthetic, the van is as sleek and curvy as they come. It is also arguably more practical. As Motor One.com reports, the roboban can hold up to 20 passengers and sports sliding doors, glass doors in the center where they can board. There is no driver as it is fully autonomous. Tesla's plans for the vehicle are extensive. Concept art shows it working as a school bus or as a pickup and drop-off shuttle at an airport. It could also be owned privately to transport a sports team, for example, or even as a pop-up roadside coffee shop or bar.

There are a few technical details, nor release date and pricing, but Elon Musk's. confirm that it will definitely enter production at some point in the future. Quote, we want the future to look like the future, he exclaimed, we're going to do this and it's going to look like this, end quote. AMD has launched its instinct MI325X GPU to compete with NVIDIA's upcoming Blackwell chips and says production will start in 2024 but didn't disclose its pricing. Quoting CNBC, if AMD's AI chips are seen by developers and cloud giants as a close substitute, for Nvidia's products. It could put pricing pressure on Nvidia, which has enjoyed roughly 75% gross

margins, while its GPUs have been in high demand over the past year. In the past few years, Nvidia has dominated the majority of the data center GPU market, but AMD is historically in second place. Now, AMD is aiming to take share from its Silicon Valley rival, or at least to capture a big chunk of the market, which it says will be worth $500 billion by 2028. A.I. Demand has actually continued to take off and actually exceeded. expectations. It's clear that the rate of investment is continuing to grow everywhere AMD, CEO Lisa Sue said at the event. AmD didn't reveal new major cloud or internet customers for its Instinct GPUs at the event, but the company has previously disclosed that both Meta and Microsoft

buy its AI GPUs and that OpenAI uses them for some applications. The company also did not disclose pricing for the Instinct MI325X, which is typically sold as part of a complete server. With the launch of the MI325X, AMD is accelerating its product schedule to release new chips on an annual schedule to better compete with Nvidia and take advantage of the booming AI chips. The new AI chip is the successor to the MI300X, which started shipping late last year. AMD's 2025 chip will be called MI350 and its 2025 chip will be called MI-400, the company said. The MI325X's rollout will pit it against Nvidia's upcoming Blackwell. chips, which NVIDIA has said will start shipping in significant quantities early next year. A successful launch for AMD's newest data center GPU could draw interest from investors that are looking

for additional companies that are in line to benefit from the AI boom. AMD is only up 20% so far in 2024, while NVIDIA stock has risen over 175%. Most industry estimates say Nvidia has more than 90% of the market for data center AI chips, end quote. Microsoft plans to let U.S. users buy and play Xbox games from the Xbox app on Android starting in November after a judge ordered Google to open up the Play Store. Quoting the Verge. The court's ruling to open up Google's mobile store in the U.S. will allow more choice and flexibility, says Xbox President Sarah Bond in a post on X. Our mission is to allow more players to play on more devices, so we are thrilled to share that starting in November, players will be able to play and purchase Xbox games directly from the Xbox app on Android, end quote.

Xbox store purchase support directly in the mobile Android app will allow Microsoft to sell Xbox games freely on Android devices and integrated Xbox cloud gaming features mean people will be able to immediately stream games to their headsets after purchasing them. It's not immediately clear why Microsoft can't already provide Xbox game purchases inside its Android mobile app, though. You can buy games from Steam or the PlayStation Store on mobile devices, so Microsoft has long been the exception here. I ask Microsoft for more details on the store changes, but,

but the company says bond statement is all it has to share right now. Separately, Microsoft is also working on a browser-based Xbox mobile store that it was originally planning to launch in July. The store will eventually focus on first-party mobile games from Microsoft's various studios, but initially it will feature deals and in-game items. Microsoft said in August that testing had begun on the web-based mobile store, end quote, work is progressing well,

and we will have more to share in the future, end quote. Remember OpenAI's GPT store, Or how's that going? Well, according to Wired, some developers say it is a mixed bag so far, with revenue sharing reserved for a tiny number of GPT creators in an invite-only pilot program that's also exclusive to the U.S., quoting Wired. In some ways, GPTs are similar to apps, though OpenAI makes a distinction between the lightweight GPTs and enterprise applications built on top of its API. OpenAI's move to create a marketplace for developers was part of its strategy to position itself as not just a chatbot maker, one of the most important platforms in the AI era.

Villicito, a 22-year-old med student in the Philippines, saw these GPTs as a way to bring in extra income for himself. They didn't require advanced coding. He ended up building more than 250 GPs. His book's GPT, which churns out personalized book recommendations and was promoted by OpenAI at the store's launch, is his most popular. But 10 months after its launch, it seems that revenue sharing has been reserved for a tiny number of developers and an invite-only pilot program run by OpenAI.

Villicito, despite his efforts, wasn't included. According to Villacito and other small developers who spoke with Wired OpenAIs, GPD store has been a mixed bag. These developers say that OpenAI's analytics tools are lacking and that they have no real sense of how their GPTs are performing. OpenAI said that GPT creators outside of the U.S. like Vilicito are not eligible for revenue sharing. Those who are able to make money from their GPTs usually devise workarounds like placing

affiliate links or advertising within their GPTs. Other small developers have. have used the success of their GPs to market themselves while raising outside funding. The AI-powered scientific research app consensus, which runs a popular GPT, raise millions in venture capital funding in August. Like Villicito, Adrian Lin, who lives in Singapore, was optimistic about the GPT store. Lin is a full-time AI researcher who has spun up an App Lab, Adrian AI Lab, on the side. He's not formally trained in web or mobile app development, so when OpenAI said that making a GPT would require no coding skills, Lynn

saw it as an opportunity to build and manage his own products. It was easy to fine-tune his models based on his intentions, Lynn says. Lynn estimates that Copywriter GPT, his GPT that drafts advertising copy had between 500 and 600,000 interactions like Villicito's books GPT. Lynn has been featured on the homepage of OpenAI Store. But Lynn can't say exactly how much traction his GPTs have gotten or how frequently they are used because OpenAI only provides, quote, rough estimates to small developers like him. And since he's in Singapore, he won't receive any payouts from OpenAI for the usage of his app, end quote. Time for the weekend long read suggestions. First up, given this week's news with all the

AI wins for Nobel Prizes, I thought this was an interesting essay. I disagree with some of it, but it's interesting nonetheless, from Bloomberg, quote. Over the past decade, Google researchers have forged some of the biggest breakthroughs in the field of AI, the T in chat GPT, OpenAI's popular chatbot is named for a technique invented in Google Labs. Translating these advances into hit products of Google's own was the problem. To remedy the situation Google last year merged its two premier AI research labs to form a new super unit Google DeepMind and fortified its ties with product teams. The lab is still working on applications of AI for climate science and medicine,

yet its primary focus is Gemini, Google's flagship AI model. It's a change for researchers, some of whom feel the mobilization around, Gemini has left less room for experimentation, as we reported earlier this year. Some of Google's biggest contributions to the field came from small self-directed teams. Google's famous paper in 2017 on machine learning, attention is all you need, had eight authors, each of whom has become a celebrity in the insular world of Silicon Valley. More than 1,200 contributors were listed on a paper for Gemini published last year. In the New World Order, some people familiar with the company's operations told us they fear pure research is getting short shrift.

And after the merger of Google Brain and DeepMind, some teams focused on scientific applications of AI were concerned their projects would be scrapped entirely. If the DeepMind folks were starting today, would they have achieved the same groundbreaking scientific work? Said Oran Etizoni, an AI researcher who founded True Media.org, a nonprofit combating political disinformation. It's a very fair question and an undeniable risk that they don't do the same level of fundamental research that we've become accustomed to, end quote. And finally, a story I've been wondering about, how is the reconstruction of the Notre Dame Cathedral in Paris going? GQ has an update, and you'd better believe there is some cutting edge tech involved in this, but also not as cutting edge, because the point is, they want to make the resurrected church as authentic as they can, and that means using older authentic techniques. quote, like in Romania, traditional French axemen used hand tools. The Haché de Grosurre, a long-handled, narrow-bladed axe used to remove large amounts of wood, and the DeLore, a broad-bladed,

short-handled axe designed for precision shipping following the grain of the wood. It's slow and physical work. Squaring lumber by hand can take much more time than doing it with buzz saws. For that reason, by the 20th century, axes had all but disappeared as a construction tool. Yet their proponents extoll the end result, more pliable, stronger beams and imperfections that reflect the extensive labor and love of craft. This is a kind of magical work because you feel the material, Foremont told me, you smell it and touch it. Calme became one of the world's connoisseurs of the Deleur, a tool that conjures images of crusaders and medieval serfs. Traveling across Eastern Europe in search of the perfect acts, Calme would delight in the differences in the shape of the cutting edge, the kind of steel, the curve of the handle. In 1992, he formed Carpenters Without Borders,

an association of master craftsmen that travel to construction sites around the world. They used newly felled trees and axes to build forges and octagonal wells in Romania, a house in China, a blacksmith shop in rural Maine. In 2020 on the lawn of an 18th century castle near Normandy coast, Carpenters Without Borders put on a demonstration for Notre Dame's decision-makers. Twenty-five volunteer carpenters built a replica of the knave's Fermé, number seven, the most technically challenging truss. Once they had joined the timbers, the carpenter slowly wasted the truss off the ground so the grandees could imagine how it would look in the church. They showed that you can do the work by hand, and it won't be much more expensive or slower,

Fremont told me, in a slight concession to modernity, they used buzz saws to square two of each log's four sides. For Philippe Jost, an engineer who was, then serving as General Gorgillens' deputy director, the display confirmed for him Fremont's instincts about how to proceed. We think it best that the cathedral remain the same, Jost told me. Truth comes through the genuine materials, the oak, the stone, and respect for the monument as we knew it." End quote. It wasn't only the roofs Sharpente that would be rebuilt according to original methods. All 460 tons of the original lead had melted in the fire and toxic particles had settled across Paris,

contaminating parks, streets, and buildings. Yet, as the medieval masters understood, lead is flexible, long-lasting, and capable of withstanding intense gusts of wind. These days, the use of lead is regulated in France, but permission was granted for the original spire's material to be used in the restoration. A replica of Violet Le Dux,

varnished oak-frame spire, coated with a lead skin, would rise above the church as well, end quote. All right, I do have a bonus episode for you this weekend. I'm going to share one of the rad 80s, 90s history episodes. Of course, I want you to go over and subscribe and listen in that feed. But if you want to taste first, I'm sharing the episode with M.G. Siegler, where we discussed the history of Blockbuster video. I figured I'd share that one because it's closest to

tech history of the ones we've done so far, you know, the whole how did Netflix kill them thing. But also, I worked at Blockbuster when I was in high school and college, so I'm able to bring some behind-the-scenes details of how the business actually worked. Enjoy that, and if you do, please subscribe to the full podcast. The plan is to have John Gruber on for the next episode talking about the making of Golden Eye 64. Talk to you on Monday.