Tech Brew Ride Home - Mon. 04/29 – AI Strategies

Starting point is 00:00:00 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 TechMeme right home for Monday, April 29th, 2024. I'm Brian McCullough today. Apple continues discussions with folks to partner for AI products. Will the upcoming iPad event kick off Apple's AI strategy? Why does spending on AI seem to work for Google and Microsoft, but not for Meta? Why were a bunch of Apple users signed out of their accounts this weekend and has AI already ruined Meta's family of apps? Here's what you miss today in the world of tech. Mark German's sources say Apple has renewed discussions with OpenAI about using their technology. to power some features in iOS 18, talks with Google about using Gemini to do similar, remain ongoing as well. So again, it seems like homegrown Apple AI tech, while there might be some of it coming to iOS,

Starting point is 00:01:21 is maybe not yet ready to the degree that Apple feels it can go it alone. Quote, the move marks a reopening of dialogue between the companies. Apple had talked to OpenAI about a deal earlier this year. The work between the two parties had been minimal since then. Apple also remains in discussions with Alphabet's Google about licensing that company's Gemini chatbot. Apple hasn't made a final decision on which partners it will use, and there's no guarantee that a deal will be worked out. It's possible that the company ultimately reaches an agreement with both OpenAI and Google, or picks another provider entirely. Representatives for Apple, OpenAI, and Google declined to comment.

Starting point is 00:01:58 The next iPhone operating system will include several new features based on Apple's in-house large language model, AI software that can generate human-sounding text. But the company also has been seeking partners to power a chatbot-like feature akin to OpenAIs chat chTPT. Bloomberg first reported in March on the discussions, which have included AI startup Anthropic as well. Relying on partners would help accelerate Apple's push into chatbots and sidestep some risks. By outsourcing the generative AI features to another company, Apple is potentially lessening the liability for its platform, end quote. More from German on Apple's AI strategy. His sources say the new iPad Pro launch coming, what is it, next week, will possibly come with an M4 chip and might serve as the unofficial kickoff of Apple's AI strategy that will be expanded upon at WWDC. Quote, the big change with the M4, a new neural engine will pave the way for fresh AI capabilities. Now here's another development. This year's Max may not be the only AI.

Starting point is 00:03:05 driven devices with M4 chips. I'm hearing there is a strong possibility that the chip in the new iPad Pro will be the M4, not the M3. Better yet, I believe Apple will position the tablet as its first truly AI-powered device and that it will tout each new product from then on as an AI device. This, of course, is all in response to the AI craze that has swept the tech industry over the last couple years. By introducing the new iPad Pro ahead of its worldwide developers conference in June, Apple could lay out its AI chip strategy without distraction. Then at WWDC, it could focus on how the M4 chip and the new iPad pros will take advantage of the AI software and services coming as part of iPad OS 18 later this year. I fully expect Apple to position the A18 chip in the iPhone 16 line

Starting point is 00:03:52 as built around AI as well. To be fair, though, these new products aren't engineered and developed entirely around AI. This is partly about marketing. Hardware with Even more impressive capabilities is further out. As I've reported, Apple is working on a tabletop iPad connected to a robotic arm, as well as a home robot. Beyond the M4 chip, the big change to the iPad Pro will be an OLED screen, bringing the CRISPR display technology from the iPhone to Apple's tablet. The iPad Air is getting a 12.9-inch screen size option for the first time, too. I've already discussed the new Magic keyboard in depth, but here's a tidbit on the latest Apple Pencil. It will have haptic feedback for the first time, end quote. Of course, one of the pressures Apple is feeling

Starting point is 00:04:41 around AI is from investors who want to see what Apple's strategy is going to be vis-a-vis this new computing paradigm. We've discussed before how Apple is in sort of a unique position. No cloud computing business like Microsoft, Alphabet, and Amazon. No huge publicly facing AI products like meta, or at least not yet. But that brings us to something interesting we saw just last week. Basically, investors cheery. Microsoft and Alphabet's huge AI spending plans as shares of both companies rallied significantly. The exact inverse happened to meta, which saw its worst trading day in 18 months. What gives here, quoting the journal?

Starting point is 00:05:21 The different reactions show how investors one year into the AI boom are becoming more discerning in how they interpret company spending and want to see a path to profit, said Rishi Jaluria, an analyst with RBC Capital. there is more scrutiny placed on what a company's AI strategy is, he said. If you don't have a strategy and are spending a lot, then there is going to be concern, end quote. Tech companies are in the midst of an infrastructure buildout not seen since the dot-com days in an expensive race to create products and services that use artificial intelligence. Building to accommodate AI is costly because the new technology requires a lot more computing power,

Starting point is 00:05:55 energy, and pricey chips. Microsoft said it is on a path to spend $14 billion a quarter on equipment and property in the midst of its AI buildout, a number that it projected would increase in subsequent quarters. The company also pointed to evidence of some early returns on its investment. The company said seven percentage points of the quarterly revenue growth in its Azure cloud computing business came from the use of its AI services. Much of that was likely from the use of OpenAI's products such as ChatGPT, which is hosted exclusively on Azure. Microsoft also charges for its AI-infused copilot assistance, including its coding assistant GitHub copilot,

Starting point is 00:06:32 which the company disclosed has 1.8 million subscribers. Shares of Microsoft and Google Parent Alphabet jumped after the company's suggested the spending was to meet growing demand for their AI products. That is partially why you see the capital investment in the shape it is, Microsoft's financial chief, Amy Hood said on the company's earnings call Thursday, because right this minute, we do have demand that exceeds our supply by a bit, end quote. As Hood detailed the company's spending plans, Microsoft shares rose as much as 5% in after-hours trading. Hood said Microsoft will spend in sync with emerging demand movements. These expenditures over the course of the next year are dependent on demand signals and

Starting point is 00:07:08 adoption of our services, she said, end quote. In other words, investors see the AI as a rising tide, rising the boats of companies with existing cloud products and infrastructures, which notably meta doesn't have. In other words, investors see AI as a tide rising the boats of companies that have existing cloud products and infrastructures. They are seeing demand right now and are spending to meet it, also already profiting on that more in a second. Meanwhile, with meta, things are still speculative. Sure, I mentioned last week they are positioned well to introduce normal users to AI for the first time, but aside from spending more time inside of meta's apps, does that make meta more money like at all?

Starting point is 00:07:54 and then there is the issue that all these sorts of AI products are so expensive to run. Quoting again, people get nervous with them and CAPEX, said one analyst of meta. Historically, Wall Street has been cautious around overheated infrastructure spending in the face of promised future profit. Some observers are comparing the current AI investment cycle to telecom companies' exuberant overbuilding of fiber optic networks more than 20 years ago. The dot-com boom of the 1990s led to speculation that soaring internet use would require enormous amounts of fiber optic cables to carry the traffic. Companies including Tyco International

Starting point is 00:08:28 spent billions of dollars laying such cables that ultimately weren't needed. The cost of that overbuilding, plus the dropping price of broadband from oversupply forced companies such as above-net, exo communications, and global crossing to seek bankruptcy court protection. The low prices and excess capacity dogged the telecom industry for more than a decade. Unlike many companies in the dot-com era, the mega-caps behind the current buildout of cloud infrastructure are already profitable, said Lake Chu, a portfolio manager at Alliance Bernstein in a recent report. Today's profitable AI stalwarts are spending on cloud infrastructure primarily to improve efficiency, end quote. Microsoft and Alphabet said much of the extra AI spending is being offset by cuts in other

Starting point is 00:09:08 parts of their companies. They said their profit margins would be only slightly affected, thanks to cost management they have done in other parts of the business, end quote. But it has been proven with the huge losses Facebook has delivered from the Metaverse and spending therein, I guess Wall Street is not as confident in them when it comes to margin discipline. Real quick, let me underline some of the numbers that we just sort of obliquely hinted at. Just this last week, Meta, Microsoft, and Alphabet disclosed that collectively they had spent more than $32 billion on data centers and other capital expenses in Q1 as they accelerate AI spending. That's just in one quarter. I saw a tweet this weekend that I have subsequently

Starting point is 00:09:49 lost track of, suggesting that this is more than ExxonMobil spends per year, not just in one quarter, but per year to extract oil from the ground. Something something, data is the new oil indeed. The Financial Times has become the latest publisher to deal with OpenAI, to train AI models on its archive content, and let ChatGPT respond with short summaries of FT articles, quoting, well, the Financial Times. agreement also allows chat GPT to respond to questions with short summaries from FT articles with links back to FT.com. This means that the chatbot's 100 million users worldwide can access FT reporting through chat GPT while providing a route back to the original source material.

Starting point is 00:10:39 Apart from the benefits to the FT, there are broader implications for the industry. It's right, of course, that AI platforms pay publishers for the use of their material. Open AI understands the importance of transparency, attribution, and compensation, all essential for us, said FT chief executive John Ritting. At the same time, it's clearly in the interest of users that these products contain reliable sources, he said. Brad Lightcap, OpenAI's chief operating officers said, our partnership and ongoing dialogue with the Financial Times is about finding creative and productive ways for AI to empower news organizations and journalists and enrich the chat GPT experience with real-time world-class journalism for millions of people around the world.

Starting point is 00:11:16 It is the fifth such deal to be struck by OpenAI over the past year, following similar agreements with the U.S.-based Associated Press, Germany's Axel Springer, Francis LeMond, and Spain's Prisa Media. Financial terms were not disclosed, end quote. Interesting to me that this specifically allows people to get summaries of FT stories. This is, I believe, the first time a publisher has allowed something like that. A number of Apple users said over the weekend they had been logged out of their Apple IDs across multiple devices. This started apparently on April 26th. that forced them to reset their passwords before logging back in.

Starting point is 00:11:59 This includes our own Chris Messina, who was out of town at an AI conference, which is kind of not when you want to suddenly be locked out of your accounts. Not that there's a good time for that to happen. Quoting 9 to 5 Mac, there appears to be an increasingly widespread Apple ID outage of some sort impacting users tonight. A number of people on social media say they were logged out of their Apple ID across multiple devices on Friday evening and forced to reset their password before logging back in. Apple's system status web page doesn't indicate that any of its services are having issues this evening.

Starting point is 00:12:30 Still, it's clear based on social media reports that something wonky is going on behind the scenes at Apple. A few of us here at 9 to 5 Mac have also been directly affected by the problem. People are being signed out of their Apple ID across all their devices. If you try to sign in with your original Apple ID password, you'll be locked out of your account. You'll then be forced to reset your password before being able to sign back in. There doesn't appear to be any rhyme or reason as to why this is happening. Whether what's happening on Friday evening has any relation to the ongoing password reset attacks problem that we've been tracking for several weeks is unclear right now. Additionally, if you have stolen device protection enabled, being signed out of an Apple ID randomly can present an especially big headache if you're away from a trusted location.

Starting point is 00:13:11 Another problem, if you reset your Apple ID password, any app-specific passwords you had previously set up via ICloud will be reset as well. I've asked Apple for more information and will update if I hear anything back, end quote. What's interesting is that, as far as I can tell, we still haven't heard anything about this from Apple, which seems odd. Finally today, kind of a product review. How is the whole putting AI into meta's family of apps going? Fast Company says AI is making meta's apps basically unusable, as the company adds meta AI everywhere, and viral user-generated AI images proliferate on Facebook and Instagram. Quote, The meta-AI experience has so far been a spam-filled one. Nowhere is that clear than on Instagram where the search function wants a place to look up a friend's account now exists seemingly to usher users into a conversation with a chatbot.

Starting point is 00:14:09 Ask meta-AI anything it now reads in my search bar? Um, no. I just want to look up my dog's daycare to see if they posted any pictures of her. This new Instagram header also features a side-scrolling list of prompts for interacting with the AI. Imagine rooftop gardens. Advice for modern stoics. anime reels. Frankly, it's bizarre that the most powerful social media advertising company in the world has so little information about me after a decade and a half of use so that it cannot even give me personalized suggestions about how I might use this dang thing. I have no rooftop, and if they think I'm interested in stoicism or anime, they're beyond help. Worse, the meta-a-I has already

Starting point is 00:14:47 been acting strangely. Seemingly unprompted, the AI chimed in by replying to a post on a parents-focused Facebook group to claim it had a disabled yet gifted child, according to 404 media. I have a child who is also 2E and has been part of the NYC gifted and talented program, the bot said, 2E or twice exceptional is a term for gifted students with intellectual disabilities. The rules of engagement on Facebook groups are complicated, the etiquette arcane, but here's one good rule for any robot cosplaying as a human aide. Don't pretend to be a sentient being with a disabled child. But to Harp exclusively on this influx of MetaMade AI would be to ignore the onslaught of user-generated

Starting point is 00:15:25 AI that's proliferating across the system. My Facebook is engulfed with AI-generated images and gullible users who seem to have no idea they are looking at fakes. One popular page called Classic Living keeps appearing in my news feed. This reminds me of Grandma's Kitchen, one post read alongside an image of an ornate greenhouse-looking kitchen with teal cabinetry and stone archways outlining casement windows, but if you zoom in, you'll see that the dangling pots and pans are disfigured. The stones are angled oddly, and the knives in the block all blend together. Still, the reviews are raves. This is an amazing kitchen, one person commented. Now that's my dream kitchen, another said. A few caught on. Don't know about you, but my grandma's kitchen was

Starting point is 00:16:05 designed and crafted by human beings, not a computer. The introduction of meta-a-I, a suite of intentional tools designed to make users' lives easier and more importantly for the company, spending more time with meta products, is an intentional salvo into a trendy area of emerging technology. If it fails, it'll either be because users genuinely do not want to interact with AI, where they're seeing friends' baby pictures, or it'll fail because it's made the entire experience kind of crappy and annoying. But the real scourge of AI is coming from the users. After years of countering misinformation around politics and the pandemic, AI has bred a kind of confusion that's hard to fully scrub or even detect.

Starting point is 00:16:42 All this AI has a confounding effect, too. It's mere purpose. presence makes people question reality even when images are in fact real. On social media, reality has never been more bent and twisted, end quote. Nothing for you today. Talk to you tomorrow.

Tech Brew Ride Home - Mon. 04/29 – AI Strategies

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.