Tech Brew Ride Home - Mon. 06/26 – The M&A Deluge In AI Begins

Episode Date: June 26, 2023

Well, this AI moment is only about 6 months old, but it looks like the mergers and acquisitions are beginning in earnest. Wait, we’re still worried about 5G interfering with airplanes? Amazon wants your local florist to deliver your packages. An Amazon/FTC kerfuffle we missed. And what are the OTHER Apple products we can expect in the next year or so? Sponsors: JoinDeleteMe.com/techmeme The Traceroute Podcast Links: Databricks Strikes $1.3 Billion Deal for Generative AI Startup MosaicML (WSJ) Pete Buttigieg Warns of Flight Delays as 5G Deadline Looms (WSJ) Exclusive: Amazon launches local business delivery network (Axios) FTC sues Amazon over ‘deceptive’ Prime sign-up and cancellation process (CNBC) Amazon, Friction, and the FTC (Stratechery) Apple Has Finally Unveiled the Vision Pro. Here’s What It’s Launching Next (Bloomberg) Learn more about your ad choices. Visit megaphone.fm/adchoices

Welcome to the Tech meme right home for Monday, June 26th, 2023. I'm Brian McCullough today. Well, this AI moment is only about six months old, but it looks like the mergers and acquisitions are beginning in earnest. Wait, we're still worried about 5G interfering with airplanes. Amazon wants your local florist to deliver your packages. In Amazon FTC kerfuffle, we missed. And what are the other Apple products we can expect in the next year or so? Here's what you missed today in the world of tech. Well, you knew the M&A activity had to start hot.

hotting up in the AI space at some point, probably sooner rather than later, Databricks has apparently agreed to acquire generative AI startup Mosaic ML in around a $1.3 billion deal, which will close in Q2. Mosaic ML launched back in 2021, raised $64 million, and has around 62 employees. Quoting the journal, Databricks, a data storage and management startup based in San Francisco, says the deal combines its AI-ready data management technology, with Mosaic ML's language model platform, enabling businesses to build low-cost language models themselves with proprietary data. Right now, most businesses rely on third-party language models trained on troves of publicly available

data access online. Also based in San Francisco, Mosaic ML, which was launched in 2021 and will become a standalone service belonging to Databricks, has been focused on bringing down the cost of using generative AI from tens of millions of dollars to hundreds of thousands of dollars per model, according to Navine Rao, co-founder, and CEO. Mosaic ML has 62 employees and has raised $64 million to date. Companies like Anthropic and OpenAI license ready-made language models to businesses, which then build generative AI apps on top of them. Driven by strong commercial demand for these models, the generative AI market has expanded dramatically, creating openings for startups like Mosaic ML that say they can offer similar AI models, but at lower.

cost and customized with a company's data. If you're building a model from scratch, you know what you're feeding it, said Databricks Chief Executive Ali Goetsey, off-the-shelf models, which are ready to use because they have already been trained on internet data, are filled with extraneous information that can skew results, Goetze said. Many companies are also wary of privacy and security issues around sharing their data in models built by outside vendors, he said. Some machine learning experts and AI vendors say that the computing and synthesis power of large language models like the one powering chat GPT, Trump's smaller models, which have powerful but ultimately limited capabilities in a specific area. Plus, ongoing challenges remain

in data management and determining which models are best suited for certain uses, said Sri Kark Krishna, KPMG's U.S. artificial intelligence leader. Corporate technology leaders are facing pressure to get their data ready for AI models. Data serves as the foundation for all algorithms because it is used to teach them to glean patterns and make predictions from it. Companies like Replit, which offers tools for programming, are already using Databricks for their data pipeline and have piped that information to MosaicML to train a code generation model, Rouse said. Known as Lakehouse, Databricks technology is designed to prepare and manage business data for AI applications, while unifying data, analytics, and AI programming tools in one system. DataBricks makes money

by renting out analytics, AI, and other cloud-based software that taps AI-ready data. Goatsy calls the picks and shovels for building enterprise tech systems. Last year, Databricks reported more than $1 billion in annualized revenue. Spending in the global generative AI market is expected to reach $42.6 billion by the end of the year and grow at a compound annual rate of 32 percent to $98.1 billion by 2026, according to market analytics firm pitch book data. It says venture funding in generative AI startups grew to $12.7 billion in the first five months of 2023 from $4.8 billion in all of 2022, end quote. Wait, what? I thought we had all this sorted out. Pete Buttigieg is warning of possible

air travel disruption ahead of a July 1st cutoff for airlines to retrofit equipment to avoid 5G interference or face restrictions. Voting the journal again. Transportation Secretary Pete Buttigieg is warning of the potential for air travel disruption ahead of a deadline for airlines to retrofit equipment to avoid potential interference from 5G wireless signals. Aircraft that haven't gone through the necessary equipment changes won't be cleared to land in certain weather conditions when visibility is low, starting July 1st, when U.S. wireless companies plan to boost their 5G service to higher power levels. There's a real risk of delays or cancellations, Buttigieg said in an interview.

This represents one of the biggest, probably the biggest foreseeable problem affecting performance this summer, end quote. So far, summer travel has been relatively. smooth, according to the journal. The severity of the potential impact is difficult to predict as it depends on weather, but it could rise to a noticeable level, Buttigieg says. He said he was urging airlines to speed up installations before low visibility conditions become more prevalent after summer or adjusts their schedules. Some airlines said they don't expect problems. Most or all of their planes now have the updated equipment, or they expect to be

able to plan around any restrictions for those that don't. The rollout of fifth-generation wireless service sparked conflict between the U.S. aviation and telecom industries as well as the government agencies that oversee them. Aviation safety officials cautioned that some 5G signals could confuse devices that use radio waves to measure planes' distance from the ground, providing readings commonly used to land in poor weather conditions. The wireless companies disputed that claim. The issue came to head early last year. Some incoming international flights were canceled as airlines scrambled to prepare for restrictions the Federal Aviation Administration planned if wireless providers began beaming the new signals to customers. In a last-minute compromise,

AT&T and Verizon agreed to delay the launch of some 5G service near key airports while carriers work to retrofit sensitive radar altimeters, and later agreed to extend that. Wireless companies worldwide have used 5G upgrades to improve their services in recent years. The standoff over the technology's air safety implications has mostly centered around the U.S. where wireless companies use different radio frequencies than their overseas peers. With the July 1st deadline now approaching after about 18 months, more than 80% of the domestic fleet and about 65% of international aircraft that fly to the U.S. have radar altimeters that won't be susceptible to interference from 5G signals, Buttigieg said, end quote.

Amazon is planning to build a local delivery network in 23 states by tapping thousands of small U.S. businesses such as bodegas and florists to deliver the packages for them. Quoting Axios. Amazon on Monday will start actively recruiting Exeos. Amazon on Monday will start actively recruiting existing small businesses in 23 states, including Alabama, Alaska, Arkansas, California, Florida, Iowa, Indiana, Illinois, Massachusetts, Minnesota, North Carolina, North Dakota, New Jersey, New York, Ohio, South Dakota, and Washington. At least 20 dense cities across the country, including Boston, New York, Los Angeles, and Seattle, will be targeted by the program. The company is interested in working with a wide range of businesses such as florists, coffee shops, clothing stores, among others. Amazon notes they don't need delivery experience to make the partnership. work. Dubbed Amazon Hub Delivery, this is the tech and logistics giant's latest attempt to expand its last mile network, the last stage in logistics where packages are ultimately delivered to customers through external workforces. How it works. Participating businesses deliver an average of 30 packages a day for seven days a week, excluding major holidays. Meanwhile, drivers from Amazon's delivery

service partner network drop off the packages to local businesses, which are required to have a secure area for storage. Amazon wouldn't state exactly how much it pays per package, but based on earnings of $27,000 a year, the rate would be about $2.50 a package, end quote. Back in 2011, when my then-girlfriend, now wife and I moved back to the city for the first time and re-encounter the whole idea of package delivery in cities, we briefly kicked around the startup idea of starting sort of a bodega business where we would buy storefronts near subway stops. and aside from the usual convenience items offered, people would pay us a monthly feed to take delivery of their packages and stuff. So either a good thing, we didn't do that or maybe a bad thing, because maybe we'd be in business with Amazon at this point.

Speaking of Amazon, one thing that fell through the cracks last week was the news that the Federal Trade Commission had filed a lawsuit alleging Amazon used dark patterns to steer customers to enroll in Amazon Prime without consent, and then, quote, sabotage their attempts to cancel. This is quoting from a CNBC piece from Wednesday of last week. The lawsuit filed in U.S. District Court for the Western District of Washington accuses Amazon leadership of slowing or rejecting changes that would have made it easier for users to cancel Prime because those changes, quote, adversely affected Amazon's bottom line, end quote. Amazon made it difficult for consumers to buy items on its site without Prime, and a button that instructed users to complete their transaction did not clearly see. state that they were also agreeing to join prime for a recurring subscription, the complaint states. The cancellation process is also difficult to navigate and designed to deter consumers from ending their prime subscription, the FTC alleged. Amazon used an internal term called Iliad to

describe the process, referencing Homer's epic poem about the Trojan War, the agency said, citing a report by insider. The complaint marks the third case the FTC has brought against Amazon in the last month. Amazon in late May agreed to pay the agency more than $30 million to settle cases alleging privacy lapses in its Alexa and ring units. The company said that it disagreed with the FTC's claims, but that it settled in order to move on from the matter, end quote. And something tells me this is destined to be settled as well. But since we miss the news, let me share Ben Thompson's analysis of this this morning. Ben says that while the FTC's Amazon case, in this case, raises fair points in isolation,

and yeah, that Iliad anecdote doesn't sound good. Prime overall is, overwhelmingly positive for users, and the FTC seems oblivious to the tradeoffs therein. Quote, set aside all the discussion above about the overall value of prime and the problem of freeloaders. This specific part of the complaint is absolutely ridiculous. Amazon's flow, at least as depicted by the FTC in their own complaint, is completely reasonable, and that's even before you start discussing the contrast with entities that let you sign up on the web, but only cancel by call. Amazon's entry into the cancellation process is clear, the flow is clear, and it's not a crime that they seek to educate would-be cancelers

as to why they might not want to cancel. This last point is important because it gets at why this complaint is fundamentally rooted in hostility to business. The reason to argue that dark patterns are bad is because customers are not sufficiently educated or capable enough to navigate a deliberately confusing interface that is driving you in a specific direction, like subscription to prime in the first place. I'm wary of the cost of government regulators getting involved in product design on a philosophical level, but I am sympathetic to the moral point. However, if you accept the premise of the previous paragraph, then it is inconsistent to complain about a company trying to educate consumers about the value they are deriving for a product

in the course of canceling that product. To put in another way, the FTC's complaint about dark patterns when it comes to signing up for Prime is rooted in the assumption that consumers lack knowledge and are easily tricked. The FTC's complaint about Amazon presenting reasons not to cancel is rooted in the assumption that consumers are already fully informed and ought to be able to accomplish their goal in as few clicks as possible. The better explanation is that the FTC is simply anti-business. This is, of course, downstream from chairperson Lena Kahn's famous law review article, Amazon's antitrust paradox. Khan, and much of the movement she represents, is intrinsically opposed to Big, and frankly, I'm sympathetic to the point. The problem with

this movement's critique, though, is that because it believes big is bad, it assumes that companies become big by acting badly. The reality of aggregation theory, though, is the opposite. On the internet, thanks to zero marginal cost in terms of serving new customers and zero transaction costs in terms of scalability, the biggest companies are those that serve customers most effectively, and leverage demand into power over supply. This is the opposite of the analog world where control of scarcity, i.e. control of supply, was the way to be dominant. The reason this matters is that all of our antitrust laws were created for the latter world, trying to apply the wrong framework to a new reality will only serve to increase costs or reduce access to people

on whose behalf, the regulator is ostensibly fighting, end quote. And finally, today in his newsletter this morning, or I guess actually yesterday morning, Mark German outlined how Apple is readying the Vision Pro to ship in 2024 and also takes a look at its Mac plans, including upgrading its lineup to M3 chips and early work on a 30-inch or larger IMac. quote, the company's 30-minute demonstrations of the Vision Pro for select members of the media and more recently developers and financial analysts showcase the most polished parts of the operating system. But in-air typing and many user interface elements aren't ready yet, and Apple still needs

to finish adapting its core iPad apps to the new Vision OS. There's plenty of work left to do on the rollout strategy as well, and how the headset will support third-party apps, 3D video content, and the eyesight feature for showing others your eyes during use. Apple also has to fine-tune how the device works with prescription lenses, and most importantly, ensure that the headset is comfortable to wear. Now that Apple has finally gone public with the device, it's roping in thousands of additional employees to figure all that out. With more people testing the headset, its shortcomings are coming under greater scrutiny. Many users are finding that the metal frame device feels too heavy after a couple hours of continuous use. Some also say that they've

experienced motion sickness, but on a more minor scale than with competing headsets. To address the weight issue, Apple has developed a second strap that goes over a wearer's head, but the company is considering selling that strap as an extra accessory rather than including it in the box. To market the device, Apple is planning to create new areas within its retail stores for demonstrations. The spaces will also let buyers choose the correctly sized bands and light seal, which keeps outside light from getting into the headset. The company is considering initially requiring appointments for purchases, something it did with the original Apple Watch in 2015. Apple has developed an iPhone app to scan customers' faces to help determine the correct size for the light seal. The company will issue that app to its retail stores for the purchase process and could also release it as an app to consumers who are shopping online.

The Vision Pro is slated to go on sale in early 2024, though early in Apple's view could mean any time until May. Sooner would definitely be better because Apple risks releasing the Vision Pro, which uses an M2 chip, around the same time as new Macs with M3 processors. It may also appear odd to consumers that the $3,500 Vision Pro's chip is inferior to those in cheaper 2024 Max. is already looking ahead to the next headset models. It has shifted some employees working on the original Vision Pro to new versions, a second generation high-end model and a lower-end headset. But let's get back to Apple's existing product categories. The company is focused on several key updates for the second half of this year and the first half of 2024. Other products may come later, but here you go. An M3 13-inch MacBook Pro, an M3 Pro, an M3 Max 14-inch and 16-inch MacBook

Pro models. New IMAX with 24-inch screens like the current models. The company is also conducting early work on an IMac with a screen over 30 inches, I'm told. New MacBook Air models, revamped iPad pros with OLED screens, new iPad Air to replace the current M1-based model. There are also a few products in early development, including a third-generation version of the AirPods Pro and new home equipment such as smart displays, as well as an Apple TV set top box with improved specifications. The next generation of the headsets probably won't begin arriving until 2025, but there will be a lot of new Apple devices to talk about until then, end quote. Always is, Mark, always is. Nothing for you today. Talk to you tomorrow.