Tech Brew Ride Home - Mon. 10/28 – Betting On The Election

Episode Date: October 28, 2024

As Robinhood rolls out smart contract derivatives around the election, a look at how smart markets have seemed to have their mainstream breakthrough this year. TikTok now has a Fediverse, open-source competitor. AI now has an opensource definition, or something? And Meta opensourced their own AI podcast making tool. Here’s what you missed today in the world of tech. Sponsors: 1Password.com/ride Links: Robinhood jumps into election trading, giving users chance to buy Harris or Trump contracts (CNBC) Billions in election bets raise the stakes of the presidential race (Washington Post) Apple Blocked From Selling iPhone 16 Models in Indonesia (Bloomberg) The fediverse is getting its own TikTok competitor called Loops (TechCrunch) We finally have an ‘official’ definition for open source AI (TechCrunch) Even With Venture Slowdown, Megadeals Grow (CrunchBase) Netflix Adds ‘Moments’ Feature Allowing Users to Easily Share Favorite Scenes (EXCLUSIVE) (Variety) Meta Releases NotebookLlama, An Open-Source Podcast Generator (Techopedia) 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 Monday, October 28th, 2024. I'm Brian McCullough today. As Robin Hood rolls out smart contract derivatives around the election, a look at how smart markets have seemed to have their mainstream breakthrough this year. TikTok now has a Fediverse open source competitor, AI now has an open source definition or something, and meta open source their own AI podcast making tool.

Here's what you missed today in the world of tech. Robin Hood says U.S. users that meet a certain criteria can now try. trade a Kamala Harris or Donald Trump contract via its Robin Hood derivatives unit and forecast X, quoting CNBC. Robin Hood jumps into a suddenly crowded politics prediction markets field that grew this month after Platform Kalshi won a key ruling against the Commodity Futures Trading Commission, which was seeking to stop the election outcome trading. The CFTC is appealing the ruling. Shortly after, interactive brokers launched various political contracts, including on the election. The election markets have generated some controversy this year with some

concerned about light volume allowing the markets to be skewed. International platform, Polymarket said last week that a single French user was responsible for four accounts spending millions to buy the Trump contract. But Polymarket said an investigation found the trading was not manipulating the market, end quote. Yes, on Polymarket, let's quote from the Washington Post. Around $2 billion in cryptocurrency has been wagered on predicting the next president on Polymarket as a Friday two weeks ago. The New York-based company says it receives tens of millions of visits per month. Terik Mansour, CEO of Kalshi, said in an interview this week that the site's user base has been doubling day over day since the U.S. District Court for the District of Columbia earlier this month

ruled against a Commodity Futures Trading Commission order banning bets on politics. The agency is appealing the decision. Advocates and fans of prediction markets argue that the financial incentives they create and the wisdom of crowds allow pricing on the exchanges to provide. accurate forecasts. Instead of stocks or bonds, traders on prediction markets buy stakes in potential outcomes that return $1 if they come to pass or nothing if they don't. In the run-up to an event, the market price fluctuates as traders weigh how likely they consider the different outcomes are and what stakes to buy and sell at different prices. Wagering on elections and other events is gaining popularity as Americans double down on sports betting in the wake of dozens of states

legalizing the past time. In 2023, the sports betting industry had nearly $120 billion revenue, a 44.5% increase from 2022, according to the American Gaming Association. But U.S. regulators are still hammering out what's allowed on prediction markets. The Commodity Futures Trading Commission has cracked down on election betting through a mix of proposed rules and court cases. The agency declined to comment on additional rulemaking, while the Kalshi investigation is pending. Polymarket founded in 2020 is the largest prediction market, but open only to non-U.S. investors, although tips on circumventing, that rule circulate online. Polymarket said it certifies that its biggest traders are not American.

The company has received more than $100 million in funding from investors, including billionaire tech titan Peter Thiel, a mentor to Republican vice presidential candidate J.D. Vance and Vitalik Buteran, co-founder of the Ethereum Crypto Network. The majority of Polymarkets, 210,000 users have lost modest amounts of money on the platform, according to the analytics site Layer Hub. And as of Friday, only 12% of users have made a profit, end quote. Indonesia has banned iPhone 16 series sales, saying that Apple has not fulfilled its 40% local content requirements and invested around $95 million below its $108 million commitment, quoting Bloomberg.

That's a road bump for Apple, which has enjoyed healthy initial sales of its flagship product in other Asian markets, such as China, while Apple ranks outside the top six smartphone brands in Indonesia is. it's a potential growth market with a young, increasingly tech-savvy population. The $1 trillion economy has over 350 million active mobile phones, much more than the nation's 270 million population, according to government data. The industry ministry said earlier in October that Apple has only invested $1.5 trillion rupee, $95 million in Indonesia below its commitment of $1.7 trillion.

Apple built four developer academies in the country in lieu of establishing a local manufacturing facility, though Chief Executive Officer Tim Cook said in April that the company was looking into the feasibility of doing so. Rival phone makers like Samsung and Zhaomi have set up factories in Indonesia to comply with the domestic content regulations introduced in 2017. Other ways to boost local content include sourcing materials or hiring workers in the country. Indonesia has a long track record of using trade restrictions to push foreign companies to produce more of their goods domestically, though to mix success. The government tightened import rules on a wide swath of products this year, leading to a shortage of items such as laptops and car tires, and causing a pile up in its ports. However, its long-running ban on the export of mineral ores like nickel have led to the rapid development of its battery sector, end quote. Say hello to Loops, a Fediverse TikTok competitor, which has begun accepting signups and plans to become open source and integrate with activity pub and seeks user donations.

Quoting TechCrunch. Still in the early stages, Loops is not yet open source, nor has it completed its integration with Activity Pub, the protocol that powers Mastodon, Pixel Fed, Fed, Fed, Fed, Fed, FED, PIRTube, and other federated apps. However, both of those efforts are in the works, and when complete, will allow loops to add another layer of social activity to the growing open social web known as the Fediverse, which now has north of 11.6 million users and over 1 million monthly active users. Massadon accounts for roughly 65% of that activity. Growth in this space has also encouraged other apps to adopt activity pub, like Social Magazine App Flipboard and Meta's Threads.

The latter is not yet fully integrated, but already has more than 200 million monthly active users. Loops, meanwhile, was developed by Daniel Supernoe, who has also created the Federated Instagram rival PixelFed. In fact, loops will run under the Pixel Fed project, according to an FAQ on its website. In addition to the eventual benefit of being open source and distributed, Loops claims it will not sell or provide user data to third-party advertisers, use your content to train AI models, or gain the rights to any content uploaded on its service. Instead, users only grant Loops permission to use their content, but will retain full ownership of their contributions, the Loops website explains.

Like many Fedaverse projects, Loops isn't reliant on investor backing. To raise funds, Loops will seek user donations via Patreon, Open Collective, and Libra Pay. It's also in the process of finalizing a grant that will sustain Loops development for one year. Aimed at users 13 and up, loops will allow you to follow other users as well as like, comment on or share their videos, but as a part of the Federated Web, the Open Social Web running on activity pub, remote users from other platforms like Macedon and PixelFed, will also be able to follow users' loops accounts and then view the videos in their home feed on those respective platforms. Those remote followers will also be able to like, comment on, or share videos if their platform

supports it. Videos published on the app will be held for moderation if the uploader has a low trust score, but trusted users will be able to skip the queue and publish immediately. The trust score is also used to hide problematic comments on posts and apply content warnings, super no notes. Other features like profile sharing or the ability for Lopes users to follow Mastodon and pixel-fed users in return are still to be announced the site notes, while sign-ups for loops are now open after a month-long countdown period,

wrapped, interested testers will be emailed when it's possible to actually start using the new app. That might take a day or so, Supernose said, due to rate limiting, end quote. Also, open source, the Open Source Initiative has released version 1.0 of its open source AI definition after years of collaboration with academia and the industry. Quoting TechCrunch. The Open Source Initiative, OSI, a long-running institution aiming to define and steward all things open-source, today released version 1.0 of its open-source AI definition. OSED, the product of several years of collaboration with academia and industry,

the OSED is intended to offer a standard by which anyone can determine whether AI is open source or not. You might be wondering, as this reporter was, why consensus matters for a definition of open-source AI. well, a big motivation is getting policymakers and AI developers on the same page, said OSI, Executive Vice President Stefano Mafuli. Regulators are already watching the space, Mafuli told TechCrunch, noting that bodies like the European Commission have sought to give special recognition to open source. We did explicit outreach to a diverse set of stakeholders and communities, not only the usual suspects in tech.

We even tried to reach out to the organizations that most often talk to regulators in order to get their early feedback. To be considered open source under the OSED, an AI model has to provide enough information about its design so that a person could substantially recreate it. The model must also disclose any pertinent details about its training data, including the provenance, how the data was processed, and how it can be obtained or licensed. An open source AI is an AI model that allows you to fully understand how it's been built. Muffoli said, that means you have access to all the components, such as the complete code

used for training and data filtering, end quote. The OSED also lays out usage rights developers should expect with open source AI, like the freedom to use the model for any purpose and modify it without having to ask anyone's permission. Most importantly, you should be able to build on top, added Mafuli. The OSI has no enforcement mechanisms to speak of. It can't pressure developers to abide by or follow the OSED, but it does intend to flag models described as open source, but which fall short of the definition. is that when someone tries to abuse the term, the AI community will say, we don't recognize this as open source, and it gets corrected, muffily said.

Historically, this has had mixed results, but it isn't entirely without effect, end quote. According to Crunchbase, even with the slowdown in venture investing recently, mega deals are still a thing. But if you think this is another story about AI, you'd be wrong. Quoting Crunchbase. Per the Crunch Base, Megadels board nearly 240 rounds of a $100 million or more have been announced or reported by reputable news outlets by U.S.-based startups. That number already tops the 210 raised all of last year. Of course, as the above chart

shows, the numbers are still way off from the aberration years of 2021 and 2022 when the venture market exploded to never seen before heights. But nevertheless, show investors growing desire to pump large sums of money into companies they believe in. The runaway leader in the clubhouse so far as sectors that have seen the most mega deals this year is the biotech health care industry. Those sectors have seen a combined $87, $100 million or more rounds. They include Zaria therapeutics, massive $1 billion plus round in April, and Radiology Partners' growth equity investment of approximately $720 million in February. Next in line is everyone's favorite AI. That sector actually saw the two largest rounds this year with OpenAI's $6.6 billion round.

raised at a post-money valuation of $157 billion this month, and X-AI's $6 billion round valuing the company at $24 billion post-money earlier this year. Cyber, fintech, and energy also have seen large slices of the mega deals pie. In fact, there have been huge rounds in a variety of sectors, including gaming, epic games, autonomous driving, Waymo, and Defense Tech, and Dural Industries. Not only has this year already surpassed last year's total of 100 million-plus rounds, but it also has already bested it in terms of $1 billion or more rounds with 11 U.S. startups raising such rounds compared to 10 last year.

Of course, last year can still claim the largest round, OpenAI's gigantic $10 billion plus round from Microsoft. The juxtaposition of venture funding falling, but massive mega deals increasing, is odd, but seems to indicate that investors in the U.S. at least are again willing to bet big on companies in which they see significant potential. It also shows biotech-slash-health care's strong pool, now in venture. Last year, those sectors saw 65 rounds of $100 million or more raised by

U.S.-based startups, but this year's total has already whizzed past that, likely due to the infusion of AI into those industries. Based on the current pace, this year's mega-deals total will likely hit close to 300, nearly a 50% jump from last year and a sign that not is all slow in the venture world, end quote. Get ready for memes originating from Netflix becoming more common, that's because Netflix has added a Moments feature on iOS, letting users save, rewatch, and share scenes across Instagram, Snapchat, and others, and plans on an Android launch soon. This is one of those things where I have to wonder, why did it take them so long to do something like this? Quoting Variety, per Netflix, with moments, members can quickly bookmark a scene while watching on their phone,

saving it directly to their My Netflix tab for easy access later. Members can also share their favorite scenes across platforms like Instagram and Facebook, creating a seamless way to engage with their favorite content and share it with friends. We're excited to add even more features in the future to help members enjoy and share their favorite Netflix moments. To promote the launch of Netflix's new feature, the streamer is debuting fresh marketing campaigns called It's So Good Monday, featuring stars including Giancarlo Esposito, Simone Biles, and Cardi B, nerding out over their own favorite scenes from Netflix movies and TV shows. shows. You know that feeling when a scene, character, or line from a show or movie just sticks with you, and you can't help but text your friends. You have to watch this. It's so good. Netflix's chief

marketing officer, Marian Lee, told Variety, whether it's sparking Halloween costume ideas or becoming a TikTok sensation, the incredible fandoms we see are a testament to what makes Netflix amazing. Our new campaign is all about celebrating these unforgettable moments, taking you back to the feeling you had when you first became obsessed with a great show on Netflix, end quote. Finally today, Meta has quietly released Notebook Lama, an open-source clone of that Notebook LM podcast generator from Google, quoting Techopedia. Similar to Google's notebook LM, which helps users with a one-click podcast generation feature. Meta's notebook Lama enables users to turn documents into conversational audio featuring to AI voices. Notebook Lama processes user-uploaded PDF files through a series of steps involving Meta's Lama 3.1 models.

Initially, the Lama 3.21B model converts the PDF into text format. The Lama 3.170B model then crafts a podcast script from the text, while a smaller Lama 3.1-8B model adds a conversational tone. Finally, META's parlor TTS tool transforms the script into audio, producing a dynamic dialogue between the AI characters. While many have shared their excitement and praise the tool's potential in early feedback, some have highlighted several limitations. Users reported on X that the audio outcome was not as smooth,

as the ones from Notebook L.M with occasional overlaps where the AI speakers unintentionally talk over each other. Meta has acknowledged these issues and in its roadmap for Notebook Lama envisions upgrades to enhance dialogue flow and sound quality. Plan improvements include using different language models for each AI character to better simulate back and forth discussions in a way that makes the audio sound human-like, end quote. Nothing more for you today. Talk to you tomorrow.