Tech Brew Ride Home - Tue. 12/17 – TikTok CEO Dines With Trump

Episode Date: December 17, 2024

On the same day that ByteDance asks the US Supreme Court to intervene, the TikTok CEO meets with Donald Trump. DeepMind’s big new competitor to Sora called Veo 2. The FTC comes down against so called “junk fees.” And on their historic raise, a deep dive look at Databricks, one of the biggest private companies in the world. Sponsors: Miro.com Links: Trump meets with TikTok CEO as video app challenges potential ban (NBCNews) TikTok Asks Supreme Court to Block Law Banning Its U.S. Operations (NYTimes) Google DeepMind unveils a new video model to rival Sora (TechCrunch) FTC issues rules requiring hotels, ticket sellers to reveal ‘junk fees’ (WashingtonPost) Waymo to begin testing in Tokyo, its first international destination (CNBC) His Startup Is Now Worth $62 Billion. It Gave Away Its First Product Free. (WSJ) Google’s Whisk AI generator will ‘remix’ the pictures you plug in (The Verge) 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 Tuesday, December 17th, 2020. I'm Brian McCullough today. On the same day that ByteDance asked the U.S. Supreme Court to intervene,

the TikTok CEO meets with Donald Trump, DeepMind's big new competitor to SORA called VO2. The FTC comes down against so-called junk fees and on their historic raise, a deep dive look at Databricks, one of the biggest private startups left in the world. Here's what you miss today in the world of tech. bite dance has officially asked the Supreme Court of the United States to block the U.S. law that could ban TikTok on January 19th. But perhaps more importantly than that, according to a source, TikTok CEO Shoozy Choo met with President-elect Trump at Mar-a-Lago. This is the more relevant news because some of the thinking is if Trump doesn't want TikTok banned, he merely has to not enforce the law, even if the Supreme Court intervenes or not.

Quoting NBC News, Trump expressed. having a, quote, warm spot for TikTok at a news conference earlier in the day, saying, quote, we'll take a look at the app and a possible ban. At his news conference Monday, Trump did not mention the meeting with Chu, but said he thought his electoral victory was in part due to his use of TikTok. I won youth by 34 points, and there are those that say that TikTok had something to do with that, he said. Trump lost voters ages 18 to 29, according to a national exit poll. Trump tried to ban TikTok in 2020 during his first term in the White House, but he was blocked by the courts. He changed his position this year, stunning some of

his supporters but pleasing others, including Republican megadoner Jeff Yass, a major investor in TikTok. Supporters of the ban include both Democrats and Republicans in Congress, and they argue that because of TikTok's Chinese ownership, it poses a threat to user privacy and to the information environment through the possible manipulation of content, end quote. As for the Supreme Court implications, quoting the New York Times, saying that the law violates both its First Amendment rights, and those of its 170 million American users, TikTok, which is controlled by a Chinese parent company, urged the justices to maintain the status quo while they decide whether to hear an appeal. Congress's unprecedented attempt to single out applicants and bar them from operating one of the most significant speech platforms in this nation

presents grave constitutional problems that this court likely will not allow to stand, lawyers for TikTok wrote in their emergency application. A three-judge panel of the United States Court of Appeals for the District of Columbia Circuit in early December, rejected a challenge to the law brought by TikTok its parent company ByteDance and several American users ruling that the measure was justified by national security concerns. The judges differed somewhat in their reasoning, but were united in accepting the government's arguments that the Chinese government could exploit the site to gain access to users' data and to spread covert disinformation. The First Amendment exists to protect free speech in the United States.

Judge Douglas H. Ginsburg wrote for the majority, joined by Judge Naomi Rao. Here, the government acted solely to protect that freedom from a foreign adversary nation and to limit that adversary's ability to gather data on people in the United States, end quote. In a concurring opinion, Chief Judge Sri Shrinavassim acknowledged that under the law's ban, quote, many Americans may lose access to an outlet for expression, a source of community, and even a means of income. Congress judged it necessary to assume that risk, he wrote, given the grave national security threats it perceived, end quote. Well, just like that, SORA has a competitor, as Deep Mind has unveiled V-O-2, a next-generation video-generating AI model that can create clips more than two minutes long in resolutions up to 4K.

It is behind a wait list as part of Video FX, but quoting TechCrunch. Notably, that's four times the resolution and over six times the duration, OpenAI's SORA can achieve. It's a theoretical advantage for now, granted, in Google's experimental video creation tool, VideoFX, where VO2, is now exclusively available. Videos are capped at 720p and 8 seconds in length. SORA can produce up to 1080p, 20-second long clips. Video FX is behind a waitlist, but Google says it's expanding the number of users who can access it this week.

Like VO, VO2 can generate videos given a text prompt, e.g. a car racing down a freeway or text and a reference image. So what's new in VO2? Well, DeepMind says the model which can generate clips in a range of of styles has an improved understanding of physics and camera controls and produces clearer footage. By clearer, DeepMind means textures and images in clips are sharper, especially in scenes with a lot of movement. As for the improved camera controls, they enable V-O-2 to position the virtual camera in the videos it generates more precisely, and to move that camera to capture objects

and people from different angles. DeepMind also claims that V-O-2 can more realistically model motion, fluid dynamics, like coffee being poured into a mug, and properties of light, such as shadows and reflections. That includes different lenses and cinematic effects, deep mind says, as well as nuanced human expression, end quote. And quoting from Engadgett. Google says the new model has a better understanding of real-world physics and human movement. Correctly, modeling humans in motion is something all generative models struggle to do, so the company's claim that VO2 is better when comes to both of those trouble points is notable. Of course, the samples the company provided aren't enough to know for sure. The true test of VO2's capabilities will come when someone prompts

it to generate a video of a gymnast's routine. And speaking of things, video models struggle with, Google says Vio will produce artifacts like extra fingers, quote, less frequently. Separately, Google is rolling out improvements to imagine three of its text image model. The company says the latest version generates brighter and better composed images. Additionally, it can render more diverse art styles with greater accuracy. At the same time, it also is better at following prompts more faithfully. Prompt adherence was an issue I highlighted when the company made Imagine 3 available to Google Cloud customers earlier this month. So if nothing else, Google is aware of the areas where its AI models need work.

VO2 will gradually roll out to Google Labs users in the U.S. For now, Google will limit testers to generating up to 8 seconds of footage at 720P, though doing so requires a $200 per month chat GPT Pro subscription. As for the latest enhancements to Imagine 3, those are available to Google Labs users in more than 100 countries through ImageFX, end quote. The U.S. Federal Trade Commission has issued long-awaited rules requiring companies to display the full price of hotels, concert tickets, and more up front, rather than hiding them as so-called junk fees, quoting the Washington Post.

The new regulations seek to crack down on bait and switch pricing practices, according to FTC Chair Lena Khan, who said the agency has received a groundswell of complaints about service charges, amenity fees, and a raft of other potentially obscure last-minute additions to consumers' bills. The FTC's rules still allow companies to impose fees so long as they are clearly displayed. The agency focused its prohibition on just the lodging and live event industries, not the fuller array of firms, from airlines to internet giants that have similarly stoked public anger over fees. While the FTC initially proposed broader regulations,

Khan appeared to narrow the agency's scope in a bid to secure bipartisan support among her five-member commission, which approved the proposal on a four-to-one vote. The lone opponent was Andrew Ferguson, a Republican whom President-elect Donald Trump intends to nominate to lead the FTC next year, end quote. People deserve to know up front what they're being asked to pay without worrying that they'll later be saddled with mysterious fees that they haven't budgeted for and can't avoid, Kahn said in a statement. The FTC's rule will put an end to junk fees around live event tickets, hotels, and vacation rentals, saving Americans billions of dollars and millions of hours

and wasted time, end quote. Government regulators have been pushing for greater financial transparency with the Transportation Department working to shed light on airline and baggage fee structures. Meanwhile, the Consumer Financial Protection Bureau has set its sights on reducing financial penalties that impact everyday Americans, particularly focusing on overdraft charges and credit card late fees. These initiatives have faced significant pushback from powerful industry players, with trade groups representing major airlines and banking institutions launching legal challenges against the proposed regulations. Their aggressive lobbying efforts have created roadblocks, potentially delaying or even threatening the financial benefits intended for consumers. The regulatory landscape could face additional turbulence as Republican lawmakers prepared to take control of both the House and Senate in January, with early signals suggesting they may work to dismantle some of these consumer protection measures. More fuel to my recent assertions, if you will.

Waymo has announced plans to test its robotaxies in Tokyo in early 2025, its first international expansion yet, partnering with taxi operator Nihon Kotsu and taxi hailing app Go. Quoting CNBC, to start, Nihon Katsu drivers will manually operate the Waymo vehicles to map key areas of the Japanese capital. Data from the manned test drives will help train the company's AI systems. Weamo will also test its robotaxies on a closed course in the U.S. built to mimic driving conditions in Japan. This is the first phase of the partnerships, which will last several quarters. Waymo told CNBC, adding that it expects to remain in Japan for an extended period. Our upcoming road trip, to Tokyo gives us the chance to work alongside local partners, government officials and community groups to understand the new landscape, Waymo said in a statement, will learn how Waymo can serve Tokyo's residents and become a beneficial part of the city's transportation ecosystem. Nihon Katsu will oversee the management and servicing of Waymo's vehicles in Japan, the companies said both the Japanese national government and Tokyo Metropolitan government

see driverless technology as a potential boon to the country's aging population, according to research by the World Economic Forum. Tokyo has designated certain areas as test zones for self-driving cars, hoping to hasten the arrival of safe driverless transportation systems there. Interestingly, the move into Japan marks the companies first into a left-hand traffic market, end quote. Data Bricks has raised $10 billion from A16Z, Thrive, and others, at a $62 billion valuation, making this among the largest VC raises in history. industry. Documents show Databricks has $2.6 billion in revenue, quoting the journal. Data scientists

that some of America's largest companies use Databricks software to analyze the large volumes of information they collect, a tool that's become even more valuable with the rise of artificial intelligence. Walgreens, for example, uses Databricks to help forecast inventory for filling prescriptions while Rivian uses it to improve the battery life of its electric trucks, end quote. Now, since we've not spoken much about Databricks before, though clearly there are one of the biggest private companies in the world at this point, I thought I'd give you a quick biographic sketch. Born in Tehran on the cusp of the Iranian Revolution, Databricks founder, Ali Goetze and his family emigrated to Stockholm, Sweden when he was five. His parents, both doctors targeted for their political views, found refuge in student dormitories, setting the stage for Ali's fascination with technical. and entrepreneurship. By fourth grade, he was already coding and running a small business

fixing his classmates' computers, an early sign of his entrepreneurial spirit. During his college years at Mid-Swedin University, a chance encounter with his business student roommate sparked a pivotal decision. The roommate's playful jab about becoming Goetze's future boss inspired him to pursue a dual major in computer science and business. This blend of technical expertise and business acumen would later prove instrumental in transforming data bricks from an academic project into a tech powerhouse. The genesis of Databricks emerged from Goetze's work at UC Berkeley's Amplab, where he and six fellow researchers developed Spark, a groundbreaking piece of code that revolutionized data analytics. Their creation shattered records processing

100 terabytes of data in just 23 minutes. This technical achievement caught the eye of venture capitalist Ben Horowitz of A16Z. The founders asked for a modest $200,000 to start the business and instead Horowitz cut them a check for 11 million, setting the stage for something much bigger, obviously. Initially hesitant to abandon academia, Goetzee worked part-time at Databricks while Eon Stosia served as CEO. The turning point came in late 2015 when Horowitz personally called Goethe to take the helm. His leadership transformation of Databricks revealed a surgical precision and decision-making, inspired by watching his parents perform medical procedures during his childhood. Under Goadzee's leadership, Databricks faced a crucial challenge, converting free users into paying

customers. A telling moment came when a potential client asked for a selfie, but balked it paying $10,000 for the software. Goadzee's response was strategic, adding premium features exclusively for paying customers and building a robust sales team targeting industry giants like Capital One and JPMorgan. His management style included sharing board presentations with all 250 employees, a practice inspired by a business case study, showing improved factory worker performance when aligned with management goals, a pivotal partnership with Microsoft materialized after Gatsy personally flew to Redmond to address concerns about their software quality, resulting in a $100 million sales commitment. By 2022, Databricks needed another strategic intervention, though, at a Las Vegas company offsite in early 2023. Gatsy unveiled his vision for profitability introducing the now-famous Conte curve tracking their progress. He implemented efficiency

measures without resorting to layoffs, instead focusing on strategic hiring and developing AI-powered automation tools like the R2D2Bot. The results were remarkable. Revenue more than doubled to $2.6 billion while significantly improving operating margins. Gutsi's deal-making prowess showed in the strategic acquisitions of Mosaic ML for $1.3 billion and tabular for approximately $2 billion outmaneuvering competitors like Snowflake. Rather than pursuing an immediate IPO, they seem to be opting for another private funding round, partly to support employee stock buyouts and associated taxes, though an IPO is still expected in the next year or two.

Finally, today, I know, I know another AI model, but this one is slightly different, so before it falls through the cracks, amidst the release of VO2 yesterday, Google also debuted WISC, an image generator that takes other images as prompts to suggest the subject scene and style and uses the new latest Imagine 3 version of their image generating. model, quoting the verge. With whisk, you can offer images to suggest what you'd like as the subject, the scene, and the style of your AI-generated image, and you can prompt whisk with multiple images for each of those three things. If you want, you can fill in text prompts too. If you don't have images on hand, you can click a dice icon to have Google fill in some images for the prompts,

though those images also appear to be AI-generated. You can also enter some text into a text box at the end of the process if you want to, add extra detail about the image you're looking for, but it is not required. WISC will then generate images and a text prompt for each image. You can favorite or download the image if you're happy with the results, or you can refine an image by entering more text into the text box, or clicking the image and editing the text prompt. In a blog post, Google stresses that WISC is designed to be for rapid visual exploration,

not pixel-perfect edits. The company also says that WISC may miss the mark, which is why it lets you edit the underlying prompts. In the few minutes I've used the tool while writing the story, it's been entertaining to tinker with. Images take a few seconds to generate, which is annoying, and while the images have been a little strange, everything I've generated has been fun to iterate on, end quote.

Nothing more for you today. Talk to you tomorrow.