Tech Brew Ride Home - Fri. 03/11 – Russia Bans Instagram And Designates Meta As An “Extremist Organization”

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 Tech meme right home for Friday, March 11th, 2020. I'm Brian McCalla today. Russia moves to ban Instagram and designate meta as an extremist organization. The EU and UK opened an antitrust probe into Google and meta about ads collusion. Stripe now supports crypto and of course the weekend long read suggestions. Here's what you miss today in the world of tech.

Starting point is 00:00:57 Believe me, I tried to think of a Yakov-Smirnov in Russia. Government bans you or something like. like that joke, but I didn't get there. Today, I learned, by the way, that Yaakov Smyranov was born in Ukraine. Anyway, Russia has moved to ban Instagram and designate META as a, quote, extremist organization. A source says that WhatsApp will not be affected as Russia thinks of it as a means of communication, quoting Bloomberg. Authorities blocked access to Facebook last week under a new media law, but the extremist designation, if approved by a court, would effectively criminalize. all of META's operations in Russia. The company's Instagram app would also be blocked.

Starting point is 00:01:45 The move comes amid increasing tension between Moscow and U.S. tech companies. Earlier Friday, the Speaker of the Lower House of Parliament called on prosecutors to investigate META after Reuters reported that the company had temporarily ceased internal restrictions on calling for violence against Russian soldiers due to the invasion of Ukraine. Russia has already banned certain social media companies like Facebook and Twitter, while tech companies have demonel monetized Russian state-sponsored media and blocked them in Europe. In light of the ongoing invasion of Ukraine, we made a temporary exception for those affected by war to express sentiments toward invading armed forces such as death to the Russian invaders. A met a spokesperson confirmed,

Starting point is 00:02:25 these are temporary measures designed to preserve voice and expression for people who are facing invasion. As always, we are prohibiting calls for violence against Russians outside of the narrow context of the current invasion, end quote. But I believe they also relaxed the rules that allowed for things like saying death to Vladimir Putin. I wonder what the odds are that these news items are related. Meanwhile, this gives me the opportunity to tell you that Google has rolled out a rapid air raid alerts system for Android phones in Ukraine. While meanwhile, DuckDuck Go, which has long touted its unbiased search results for years, says it will downrank sites, quote, associated with Russian disinformation, end quote. Again, this is actual downranking,

Starting point is 00:03:14 quoting and gadget. In a tweet, founder and CEO Gabriel Weinberg wrote that the privacy focused search engine would be releasing updates that would ensure Russian disinformation sites rank further down in search results. Earlier this month, Duckducko announced it would pause its relationship with Russian state-owned search engine Yandex. Even if it's for the right reasons, putting a thumb on the scales of search results is an uncharacteristic move for the Pennsylvania-based search company. Unlike Google and major social media platforms, DuckDuckGo, has neglected to filter out misleading content on vaccines or elections. A number of platforms, including the meta-owned Facebook and Instagram, have started to demote posts from Russian state media. Google has been downranking

Starting point is 00:03:55 search results from Russian state news agencies since 2017. Known for being a privacy-minded search engine Duck. Dock Go does not track its users or sell data to third parties. The company primarily makes money from affiliate links and non-targeted ads. Duck. Go, which regularly donates to digital rights groups like the Electronic Frontier Foundation and nonprofit newsroom, the markup, has also been embraced by conspiracy theorists and far-right groups in recent years. A number of studies highlighted in the New York Times found that Bing's search algorithm. which powers Duck. DuckGo surfaces more sites that promote conspiracy theories than Google. Many fans of Duck. DuckGo criticize the search engine for its decision on Russia, liking it to

Starting point is 00:04:36 censorship. It's unclear whether Duck. Go will make a wider effort to downrank disinformation, end quote. The EU and the UK have opened an antitrust probe into Google and meta's Jedi Blue deal, which related to online ads. But here's a wrinkle. The EU is also probing whether Google acted without meta's knowledge. Quoting the F.T. The move follows U.S. antitrust investigators who are also probing an agreement informally known as Jedi Blue. The search engine giant and Facebook's parent company have been accused of working together to carve up advertising profits acting together to buttress their businesses.

Starting point is 00:05:19 Companies found in breach of EU law standaloo's up to 10% of global revenues, but the legal processes could take years. Google said that, quote, the allegations made about the Jedi Blue agreement are false. It added, quote, the goal of this program is to work with a range of ad networks and exchanges to increase demand for publishers' ad space, which helps those publishers earn more revenue, end quote. Meta said its, quote, non-exclusive bidding agreement with Google and the similar agreements we have with other bidding platforms have helped to increase competition for ad placements, end quote. Both groups vowed to cooperate with regulators. Margath Vestager, the EU's

Starting point is 00:05:56 competition chief said the European Commission is also investigating the suspicion that Google may have acted alone without Meta's knowledge. Quote, we have not concluded yet if it's a Google thing alone or if they were in it together. It's not a given that Meta was conscious of the effects of the deal, and that's what we have to investigate, she said. The Commission and CMA are assessing to what extent meta abused its dominant position in the social media or digital advertising markets through its use of data. That probe is ongoing. Andrea Koskelly, CMA chief executive, said, quote, we're concerned that Google may have teamed up with meta to put obstacles in the way of competitors who provide important online display advertising

Starting point is 00:06:34 services to publishers, end quote. Stripe now supports crypto businesses, including crypto exchanges, wallet providers, and NFT marketplaces, and is now offering flexible on-ramps for crypto exchanges. Quoting the block. According to a statement on its website on Thursday, Stripe now offers Fiat Payments API integration for crypto businesses to process crypto-to-fiat currency payments. As part of the toolkit for crypto firms, Stripe says it is also offering flexible on-ramps for exchanges. Those on-ramps, as co-founder John Collison put it on Twitter, will enable both

Starting point is 00:07:15 fiat deposits and withdrawals for crypto exchanges. Stripe's crypto business toolkit also includes protocols for user identification and fraud prevention. Cryptoexchange FTX and its U.S. affiliate, FTX.U.S., have partnered with Stripe to improve their no-eastern your customer protocols. The partnership will see Stripe build user onboarding protocols including identity verification processes. According to an announcement, the partnership will also see Stripe providing payment services for the exchange. Based on the partnership, FDX and FTX users will be able to buy crypto using debit cards and direct bank transfers via automated clearinghouse or ACH transactions. Today's announcement is the first major crypto-related news from Stripe since it announced the

Starting point is 00:07:58 creation of a team dedicated to Web3 pay. in October last year, end quote. It gives me absolutely zero joy to report this, but I do feel like this is something we need to keep our eyes on. According to the Financial Times, after a peak of $248 million in February, daily trading volume of NFTs on OpenC fell 80% to around $50 million in March. The average price of a board ape NFT has fallen some 44% since February 24, quote, The average selling price of an NFT has dropped more than 48% since a November peak to around $2,500 over the past two weeks, according to data from website Non-Fungible.

Starting point is 00:08:45 Daily trading volumes on OpenC, the biggest marketplace for NFTs, have plummeted 80% to roughly $50 million in March, just a month after they reached a record peak of $248 million in February. Meanwhile, the number of accounts buying and selling NFTs on a weekly basis has fallen to about $194,000, according to non-fungible. The number of accounts hit a peak of 380,000 last November. By the end of last year, quote, there was a general sense that there was saturation in certain parts of the market, particularly in primate-themed profile pictures, said Nadia Ivanova, chief operating officer at a trend forecasting unit of the French bank BNP Paribas. I think that many will be scarred and burnt by this market and may never touch NFTs again,

Starting point is 00:09:31 said a 19-year-old investor in a telegram messaging group where more than 1,000 people discuss NFTs. Other members joke they would be living off rice, porridge, and grass this month. One blue chip NFT index offered by Bitwise has fallen 25% in the past month, leaving it down 17.1% for the year. Bored Apes and Cryptopunks, two of the most popular and richly valued collections, made up more than 60% of the index as of this week. The market's pullback has mirrored a broader sell-off in ether, the dominant cryptocurrency used to purchase NFTs, which has fallen by more than 40% from an all-time. high in November. Many projects in decentralized finance and other areas tied to Ethereum have also

Starting point is 00:10:10 plummeted in value, end quote. Time for the weekend long read suggestions. And first, this can be interpreted a couple different ways, but this piece in Wired says that two years into the pandemic, US tech jobs remain concentrated in a handful of coastal hubs. But a new set of cities is gaining ground. So I would say to that, look, no one said Silicon Valley was going to go away. I just 100% believe the sentiment around where tech can be and thrive has sort of expanded probably forever. Quote, over the past few decades, high-paying tech jobs in the U.S. have increasingly concentrated in a handful of cities contributing to regional economic inequality. The tech sector grew by 47% in the 2010s. And in the latter half of that decade, nearly half of tech job creation,

Starting point is 00:11:06 in eight superstar metro areas, San Jose, New York, San Francisco, Washington, D.C., Seattle, Boston, Los Angeles, and Austin, Texas. By the end of the decade, those eight cities comprised 38.2% of tech jobs. Located largely in the nation's interior, nine rising stars, Atlanta, Dallas, Denver, Miami, Orlando, San Diego, Kansas City, St. Louis, and Salt Lake City had increased their share of tech jobs before the pandemic struck, growing by an average annual rate of 3% between 2015 and 2019. Like the superstars, these cities boasted proximity to large universities and an abundance of highly educated technical workers. Pandemic-driven remote work did little to loosen these cities' stranglehold on jobs.

Starting point is 00:11:49 In 2020, the pandemic's first year, both superstars and rising stars added tech jobs, slightly increasing their overall share. The rate of growth, however, slowed, dropping from about 5% pre-pandemic to 2.9% in 2020. Instead, 36 other cities notched stronger tech job growth than before the pandemic. These included northern business centers like Philadelphia and Minneapolis, large warm weather cities like Charlotte, North Carolina, big university cities like Chapel Hill, and vacation centers like Virginia Beach. Ammany rich and vacation towns such as Santa Barbara, Barnsville, Massachusetts, saw job growth

Starting point is 00:12:24 surge by more than 6 percent, while college towns such as Boulder, Colorado, and Lincoln, Nebraska grew by more than 3%. And quote. Then AI guru Gary Marcus, one-time guest on this very podcast, has an essay up that's getting a lot of chatter because he's suggesting that deep learning is hitting a wall. This is slightly beyond my ken in terms of knowing the veracity of all this, but if this is your realm, you might want to read, quote, there are serious holes in the scaling argument. To begin with, the measures that have scaled have not captured what we desperately need to improve. Genuine comprehension. Insiders have long known that one of the biggest problems in AI research is the tests or benchmarks that we use to evaluate AI systems. The well-known Turing

Starting point is 00:13:09 test aimed to measure genuine intelligence turns out to be easily gamed by chatbots that act paranoid or uncooperative. Scaling the measures Kaplan and his open AI colleagues looked at about predicting words in a sentence is not tantamount to the kind of deep comprehension true AI would require. What's more, the so-called scaling laws aren't universal laws like gravity, but rather more observations that might not hold forever, much like Moore's Law, a trend in computer chip production that held for decades, but arguably began to slow a decade ago. Indeed, we may already be running into scaling limits in deep learning, perhaps already approaching a point of diminishing returns. In the last several months, research from deep mind and elsewhere on models even larger than GPT3 have shown that scaling starts to falter on some measures, such as toxicity, truthfulness, reasoning, and common sense. A 2022 paper from Google concludes that making GPT3-like models bigger makes them more fluent, but no more trustworthy. Such signs should be alarming to the autonomous driving industry, which has largely banked on scaling rather than on developing more sophisticated reasoning. If scaling doesn't get us to safe autonomous driving,

Starting point is 00:14:18 tens of billions of dollars of investment in scaling could turn out to be for not, end quote. At the same time, let me introduce. you to Transformers. Perhaps the next big leap in AI development, quoting from Quanta magazine. Imagine going to your local hardware store and seeing a new kind of hammer on the shelf. You've heard about this hammer. It pounds faster and more accurately than others, and in the last few years, it's rendered many other hammers obsolete, at least for most uses. And there's more. With a few tweaks and attachment here, a twist there, the tool changes into a saw that can cut at least as fast and as accurately as any other option out there. In fact, some experts at

Starting point is 00:14:56 the frontiers of tool development say this hammer might just herald the convergence of all tools into a single device. Well, that's the story playing out among the tools of artificial intelligence. That versatile new hammer is a kind of artificial neural network, a network of nodes that learn how to do some tasks by training on existing data called a transformer. It was originally designed to handle language, but it has recently begun impacting other AI domains. Transformers quickly became the frontrunner for applications like word recognition that focus on analyzing and predicting text. It led to a wave of tools like OpenAI's generative pre-trained Transformer 3, GPT3, which trains on hundreds of billions of words and generates consistent new text to an unsettling degree.

Starting point is 00:15:41 The success of Transformers prompted the AI crowd to ask what else they could do. The answer is unfolding now as researchers report that Transformers are proving surprisingly versatile. In some vision tasks like image classification, neural nets that use transformers have become faster and more accurate than those that don't. Emerging work in other AI areas like processing multiple kinds of inputs at once or planning tasks suggest transformers can handle even more, end quote. And then let's do a couple of profiles. The first from Fast Company covers Canva, the $40 billion design juggernaut, quote, It was May 2012, and the Australians, both in their mid-20s, had flown from Sydney to Silicon Valley six months earlier, to raise money from investors on Sand Hill Road. The couple had met as teens

Starting point is 00:16:28 at the University of Western Australia in Perth. Neither had formally studied the discipline of design. Perkins had majored in communications, Obrecht in education, but Perkins had tutored other students in Photoshop in her spare time and had become fascinated by tools that made design more accessible to, well, self-described Luddites like Aubrecht. Together, they had launched Fusion Books, an online yearbook design product that was chugging along back home. They'd come to California with a simple pitch. They needed funds to build a platform that would allow anyone to design just about anything, all within a single interface. Instead, they received more than 100 rejections. That afternoon, the final name on their once-promising spreadsheet of VCs had turned red.

Starting point is 00:17:09 So the usually optimistic pair went on a misery walk, recalls Abricht, to decide to if they had run out of options. After three hours of soul searching, they determined they hadn't. We got up the next morning and just started back at it, he says. Within a few months, they raised the first part of what would become a $3 million seed round, and a year later, they launched the online design publishing tool, Canva. Just over a decade later, Perkins and Aubrecht are married, and back in Sydney. As for Canva, where Perkins is CEO and Aubrecht is chief operating officer, it's now one of the most influential design platforms on the planet, with 75 million monthly active users who create on average 150 designs every second. The company, which is valued at $40 billion,

Starting point is 00:17:48 expects to exceed $1 billion in revenue in 2022, end quote. And then finally, the Generalist has a deep, deep dive on the history of Angel List, the company that makes the ride home fund even possible, and which just announced a $100 million series C at a $4 billion valuation. quote, when I asked Kevin Laws what he thought Angelus might look like in a decade, he said, I have no clue. That's kind of the good part of it, end quote. He has a point. How could anyone predict what this business might become, given how much it has changed in its first 12 years of existence? In a draft of a formal announcement, I reviewed Angelus Venture shared some updated figures. Close to 200 unicorns raised money from Angelus GPs, with 19 businesses entering the

Starting point is 00:18:35 public markets, a demonstration of the deal quality on the platform. In 2021, 38,000 LPs, backed by 800 GPs, just like me, resulted in $3.6 billion invested in 7,000 startups. That's a deployment rate of which Tiger itself would be proud. Tiger was the one that led the round in Angelus recently, by the way. For a company that last announced around in 2015, Angelus Ventures financing carries more meaning than most. Not only is it an indication that the company is on a breakout trajectory, but that we can expect big moves in the coming years, end quote. So the bonus episode this weekend is something I've been meaning to do for a while, and then yesterday, A16Z reposting that episode I was on years ago reminded me to do this. Chris Messina has been co-hosting with me

Starting point is 00:19:35 on the show for about a year now, but while we sometimes act like you should all know who Chris Messina is, what he's done. Maybe we can't assume that. So I'm reposting my 2019 interview with Chris that I did for the internet history podcast. Come here, Chris and I meet in person for the first time, but stay to learn about the invention of the hashtag and Chris's time in the early web 2.0 scene, working at Firefox, working at Google, working at Uber. I think after listening to this, you'll get a better sense of Chris's bona fides and why he comes at some of the topics we discuss the way he does. You'll get a sense of Chris as a person and also some of his priors when it comes to product and tech. So enjoy that tomorrow. Talk to you on Monday.

Tech Brew Ride Home - Fri. 03/11 – Russia Bans Instagram And Designates Meta As An “Extremist Organization”

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