On Our "Virtual Route 99" (Special Month End Edition): On Human Values

 We hope to do our part to overcome!!

On Our "Virtual Route 99" (Special Mid-Week Edition ): On Humanity, AI, Power As Yuval Noah Harari Reflects & the Information Updates...

 

The Briefing

By Martin Peers

You might be shocked to learn there is a world outside AI! And we got a glimpse of it in an Oakland, Calif., courtroom today, at the opening of the trial in a lawsuit brought by California and several other states against Meta Platforms over the alleged harms its social media apps have done to kids. This case, part of a deluge of such lawsuits against Meta and other social media companies that started going to trial in January, may be existential for the companies involved, given the damages sought and the operational changes requested of them. Meta, for instance, has said that the states in the Northern California case are seeking $1.4 trillion in penalties—equivalent to its market capitalization—although the states put the number at closer to $200 billion, according to Reuters. Let’s assume Meta is being hyperbolic. The point is there are thousands of cases, many filed by individuals rather than states, so the costs could quickly add up. As Meta’s finance chief, Susan Li, said on an earnings call last month, the trials scheduled for this year in the U.S. “may ultimately result in a material loss.” When Meta uses the word “material,” you know it’s serious. And look, it’s hard to dismiss the worst-case scenarios, given that Meta has already lost a couple of cases, including one in California brought by a young person, which was tried over the winter, and one in New Mexico that resulted in a total of $942 million in penalties levied against Meta. The Northern California case is more akin to the New Mexico lawsuit, as it’s brought by states rather than individuals. Both cases broadly make similar allegations: that Meta designed its apps to addict young people. As California’s deputy attorney general said on Tuesday, Meta’s business model was designed to “hook the users, hold them for as long as they can, harvest their data and then hide the truth from the public.” Things are never as simple as they appear, however. For one thing, Meta and other social media firms have so far lost attempts to defeat the lawsuits by pointing to the protection they enjoy under Section 230 of the Communications Decency Act, which gives them immunity from liability for what their platforms publish. You can foresee a scenario in which an appeals court reverses the verdicts found so far on exactly those grounds. (So far, however, one appeals court has not done so.) Leaving aside that issue, Meta also has some defenses. After all, as one Meta lawyer pointed out on Tuesday, the Facebook app is overwhelmingly used by adults, not children. (It’s no secret that the predominance of very old people has put kids off Facebook in recent years.) Instagram skews younger than Facebook, but neither of these apps is like Snapchat, for instance, which is better known for its appeal to kids. That reality might undercut the claim that Meta designed its apps to trap kids. Still, that’s the bull case for Meta. The bear case is that the New Mexico verdict demonstrates how much evidence the states can marshal against Meta that won’t sit well with the jury. We’ll have to see how this plays out. Anthropic Goes for Supervotes This is going to be interesting. The Information scooped the news on Tuesday that Anthropic was preparing to give its co-founders, including CEO Dario Amodei, a class of stock with additional voting power—typically known as supervoting shares—to give them more ability to deal with “outside shareholder pressure.” Supervoting stock is common in tech and in parts of the media sector. Theoretically it allows founders to take a long view in their management of a company rather than being beholden to very short-term–oriented investors. The theory is that without these protections, a founder-CEO could be vulnerable to an activist investor unhappy with their management and wanting a quick fix. In practice, that means founders armed with supervoting shares sometimes do things that are unpopular with investors in the short term. Meta CEO Mark Zuckerberg, who has supervoting shares giving him control of his company, is a good example: He’s spending a fortune on AI right now, which has not exactly enhanced his standing on the street. The same is true for Snap CEO Evan Spiegel and his quixotic pursuit of augmented reality glasses. (An activist has tried to persuade Spiegel of the error of his ways, so far unsuccessfully.) One issue with Anthropic, though, is that the company already has a trust that has the power to elect a majority of its board. The trust is meant to “address the unique challenges and long-term opportunities we believe transformative AI will present,” as Anthropic has put it. Our story said the new voting rights would “balance the trust’s powers”—but will they simply curtail them? The devil, however, will be in the details. In Other News • Shares of Swedish “buy now, pay later” firm Klarna plunged 20% on Tuesday after the company revised down its projection for full-year business, citing weakness in Germany, its biggest market. At the same time, Klarna announced that its longtime finance and marketing chiefs would leave early next year. • Apple appeared to inadvertently confirm the existence of a new AirPods product with cameras through an unlikely channel: the beta release of a new Mac operating system. Apple news outlet MacRumors uncovered a short video clip included with the new operating system that shows a man holding a book up to the AirPods so they can identify and record its title. “With visual intelligence, your world becomes savable,” a voice-over says in the clip. • Tesla teased a launch event for its Cybercab on Monday night. That would mark the company’s public debut of its first vehicle designed without a steering wheel or brake pedal and intended for its autonomous ride-hailing service.  Anthropic Now Earns 60% More Than OpenAI Anthropic is heading for a stock listing on a run rate that grew sevenfold this year.Anthropic's annualized revenue run rate reached $65 billion at the end of July, up from $9 billion in December. OpenAI's comparable figure doubled to $40 billion over the same stretch, while Anthropic's multiplied more than seven times.  ByteDance cuts Hollywood's first AI video deal ByteDance agreed to a formal framework with Hollywood’s Motion Picture Association (MPA) that will implement film and TV copyright protections into its Seedance and Seedream models, months after it received a cease-and-desist notice. A viral Seedance 2.0 clip of Tom Cruise fighting sparked the initial legal feud, which marked the MPA’s first cease-and-desist against a major AI company.

On Our "Virtual Route 99" With a Special #OutsiderVibes Edition: SPotlight Starring Michael Keaton

On Our "Virtual Route 99" (W-End Edition): On the True Art of the Possible As #BruceLee Reminds Us...

 

On Our "Virtual Route 99" (Special Friday Edition): #RandomThoughts For the Week

 

On Our "Virtual Route 99" (Mid-Week Edition): Looking to 2050

On Our "VIrtual Route 99 (Weekly Edition): On AI Featuring @KalToons & Thoughts By Goldman Sachs

Global AI Investment Is Forecast to Exceed $1 Trillion in 2026 The most common estimate for investment in artificial intelligence (AI) is the projection for capital expenditures by US hyperscalers. These tech companies are forecast to spend about $800 billion this year, according to the consensus of analyst estimates. But those estimates have several drawbacks, according to Goldman Sachs Research. They do not include investment in AI by private companies and by companies outside the US. Not all capex by hyperscalers is necessarily related to AI. And the major US tech companies operate globally, indicating some of their investment takes place outside the US. Goldman Sachs Research adjusted the widely used measures of US hyperscaler capex to produce a more comprehensive estimate. That projection points to $1.019 trillion of AI-related investment around the globe in 2026, including $581 billion in the US, writes Joseph Briggs, who co-leads the Global Economics team, in a report.

Goldman Sachs Research’s augmented estimates indicate that the commonly cited forecast for hyperscaler capex of $794 billion likely understates the total amount of global AI capex by around $200 billion. At the same time, that forecast likely overstates the amount of US investment by $200 billion. Read the full article for more on Goldman Sachs Research’s adjusted estimates for AI investment and its outlook for AI capex.

Quoted: The Evolution of the Modern Space Economy “Space is becoming a new pillar of the industrial economy, with supply chains, infrastructure nodes, and value pools that resemble other sectors. How companies finance themselves has evolved accordingly: Rather than relying solely on government programs, venture rounds, or strategic investors, space companies are increasingly turning to public equity markets to fund growth.” —George Lee and Dan Keyserling of the Goldman Sachs Global Institute In their latest white paper, Lee and Keyserling explore the technological innovations—from smaller satellites to reusable rockets—that are reshaping the economics of the space sector, as well as what the expanding space industry means for geopolitics. In case you missed it: Also listen to our podcast on the new commercial opportunities arising from a growing space economy.

On Our "Virtual Route 99" (Special W-End Edition)

 

On Our "Virtual Route 99" With #RandomThoughts For the Week

 

On Our "Virtual Route 99" Around the Tech Scene This Week

We present a snapshot of the latest on the tech & AI scene for all to review, courtesy the team at The Information: 

Thanks for reading The Briefing, our nightly column where we break down the day’s news. If you like what you see, I encourage you to subscribe to our reporting here. Greetings! A flood of June-quarter earnings updates—from the likes of Uber, DoorDash, Shopify and Disney—was interrupted on Thursday by Google’s announcement that its AI guru, Google DeepMind CEO Demis Hassabis, had been elevated to chair of the AI unit and chief scientist of Google’s parent company, Alphabet. That news coincided with the revelation that several others from DeepMind, including Chief Scientist Jeff Dean, were leaving to start their own firm. Some interpreted the shake-ups as bad news for Google, whose stock fell 4%. In reality, today’s news doesn’t mean much. Hassabis wasn’t the one running DeepMind on a day-to-day basis. That was really Koray Kavukcuoglu, DeepMind's Chief Technology Officer, who is now formally taking the reins of Google DeepMind. (For more on Kavukcuoglu, see this profile we published a year ago.) Hassabis, in contrast, is the public face of Google DeepMind as well as a visionary scientist, roles he will continue to perform (he’s a Nobel Prize winner, don’t forget). As for Dean, my colleague Erin Woo tells me the chatter internally is that Dean had become less central to Google’s Gemini efforts, even if he remains revered by employees. The real question about today’s shake-up is whether it changes the list of possible successors to Sundar Pichai, CEO of both Google and Alphabet. Hassabis has been on that list—at least in the minds of smart outsiders—as has Google Cloud chief Thomas Kurian. You might say that if Hassabis prefers the role of scientist to that of divisional CEO, he’d hardly want to be Alphabet’s CEO. He will continue to run Isomorphic Labs, Alphabet’s drug discovery unit, but that’s consistent with the idea that he wants to be a scientist. In other words, today’s news is good for Kurian’s chances.  That said, the question of Pichai’s successor isn’t one that has to be decided any time soon. Pichai is riding high right now. While he has lately faced investor pressure on Google’s aggressive capital expenditure plans, and questions about its latest AI models have circulated, those are short-term quibbles. Stepping back, Pichai has answered his critics by demonstrating that Google can be a leader in AI technology. He’s also rebutted worries that AI would undermine its search cash cow, and meantime cloud is soaring. Pichai is in his early 50s, which means he could be in the job for years to come.  But company boards always need options. Some might argue that Hassabis could still be in the running given that the chief of a company as big as Google doesn’t necessarily make daily operational decisions. The CEO’s job is to plot the company’s course for the future—to ensure it has a future. But the CEO still has to get their hands dirty in operational decisions. Other parts of the job are not exactly fun—remember Pichai’s appearance before Congress a few years ago? For someone who’s more interested in “actively shaping the future” of AI, as Pichai said of Hassabis today, the CEO job is likely not in his future. Maybe he wants to win a second Nobel Prize instead.  Figma’s Slump Uh-oh. Shares of design firm Figma dropped 15% in after-hours trading on Wednesday, after the company projected a sharp slowdown in revenue growth and a lower profit margin for the third quarter and announced that a couple of longtime executives were leaving.  After reporting 48% growth in the second quarter, 2 percentage points faster than in the first quarter, Figma projected growth of 36% for the third quarter. That’s still strong, but a 12 percentage point slowdown is not to be ignored—particularly at a time when AI-fueled competitors are appearing everywhere (including AI powerhouses like Anthropic). Moreover, Figma projected an operating profit margin of 9% for the full year, compared to 13% for the first six months of the year.  Figma’s explanation focused on the projected profit margin: The company is investing. It has created new products it is still testing and isn’t requiring customers to spend money on. So Figma will have to see how those products mature before it makes them “generally available.” Investors weren’t persuaded. In Other News • Shopify stock jumped 18% on Wednesday morning after the e-commerce software service reported better than projected 34% revenue growth for the second quarter. • Walt Disney Co.’s entertainment streaming service grew 11% in the June quarter, the company reported Wednesday, just a couple of percentage points behind industry leader Netflix. But Disney’s results showed that growth is coming almost entirely from the subscription side rather than advertising, a sign of how competitive the streaming ad market has become. • Uber CEO Dara Khosrowshahi said the company expects to invest $10 billion in “the coming years” to help pay for adding self-driving cars to its service, both through investments in autonomous vehicle firms and expanding infrastructure for autonomous vehicles such as managing fleets of cars. Uber also reported 12% higher revenue in the second quarter as the volume of its business in both delivery and ride hailing surged.