← Tech Brew Ride Home
Tech Brew Ride Home
Watermarking AI
00:19:00
Anthropic said new Claude models will watermark generated text to satisfy the EU AI Act, and it's out courting investors for a possibly record-breaking IPO. Apple's glass iPhone stayed on track, YouTube doubled its monetization bar, and FlightAware sued Kalshi.
Loading summary
Transcript1 lines
- [00:05]
A
Welcome to the Tech we write home for Tuesday, August 11, 2026. I'm Brian McCullough. Today, Anthropic said new Claude models will watermark generated text to satisfy the EU AI act, and it's also courting investors for a possibly record breaking IPO. Apple's Glass iPhone stayed on track, YouTube doubled its monetization bar, and FlightAware is suing Kalshi. Here's what you missed today in the world of tech. Every day shareholders meet to discuss important matters about the companies you invest in. Now you can make your voice heard too. Vanguard investor choice makes it easy to set your proxy voting preference for your eligible Vanguard index funds. Whether you hold a Vanguard fund directly or through another brokerage firm, all it takes is a few clicks to select your proxy voting preference and be heard on important shareholder topics like executive pay and director elections. Visit vanguard.com investorchoice to learn more. It's your shares. It's your voice. It's easy. Vanguard investors own shares of our index funds and those funds own shares of the companies they invest in. Vanguard Marketing Corporation distributor Anthropic says that going forward, CLAUDE models will embed watermarks not in pictures. CLAUDE doesn't really do pictures, but inside of actual generated text on also C2PA metadata and files. So solving the problem mentioned yesterday around cheating? Well, maybe, but mainly to comply with the EU AI Act. Quoting the register, generated text will carry embedded watermarks and generated files will include digitally signed provenance metadata where supported, the company said in a help document published on Monday. The AI biz is also working on models it has already released to add output marking during the transition period allowed under EU law. Marking will apply to output from supported models wherever CLAUDE is offered worldwide, the company said. But this move may further amplify the appeal of open weight models and alienate CLAUDE customers who don't necessarily want consumers of their AI generated content to know its provenance. In the past, CLAUDE users have expressed frustration with pricing changes, reliability issues and model safeguards that have hindered legitimate work. In the name of safety clothes, CLAUDE users appear to be skeptical that a text based watermarking scheme will work. Researchers have already demonstrated that image based watermarking can be undone. Anthropic intends to apply marks to output from covered CLAUDE models on the CLAUDE Platform API, CLAUDE CLAUDE code, CLAUDE cowork and CLAUDE Tag. This also includes third party providers of anthropic models like aws, Google Cloud and Microsoft Foundry. The AI biz expects to provide details about how people can detect Claude's marks as required under EU law in forthcoming documentation. When a supported Claude model generates text, it weaves an imperceptible watermark directly into the text itself, the biz said. You won't see it, and it doesn't change the meaning, quality or reliability or readability of Claude's response. Because the watermark is part of the text, it will travel with the text when it's copied and pasted elsewhere, and may persist through some editing. Watermarking will be applied at the model level, which means it will be present no matter which clod product or surface the text comes from, absent examples or technical documentation. It's unclear how Anthropic will make the watermarks hard to remove, but it should not be difficult to create an optical character recognition system that strips or omits obscure marks from Claude generated texts. Anthropic's insistence that its text marking scheme doesn't change the meaning of Claude's output appears to preclude using word choice and placement as a text provenance identifier, a technique Apple has reportedly used to catch those who would violate its employee secrecy agreements. As for attached to Claude created files, Antopic says it will rely on signed provenance metadata that conforms to the C2PA standard, for which there are already open source removal tools. End quote. The journal says the Anthropic IPO is still coming, and it could be the biggest one yet. Bigger than even SpaceX. Anthropic is meeting with potential investors to shore up confidence in what could be the largest IPO of all time, offering assurances about its rapid pace of growth and strategies it expects will help address a growing public backlash against AI generally. The $965 billion artificial intelligence juggernaut is staring down a host of fresh challenges that are leading investors to take a more critical look at the company's business, including the recent popularity of cheaper AI systems made in China, tensions with the Trump administration, and a growing backlash to data center construction across the US in recent weeks, investors have pressed Anthropic executives on such topics in pre IPO meetings, asking them what effect they might have on the company's growth. People familiar with the discussion said. The conversations reflect the tremendous uncertainty that still exists about who will win the AI race and the financial stability of the business that underpins it. Company executives have played down the impact of Chinese competition in meetings, telling investors that Anthropic is hyper focused on offering cutting edge AI models, the people said. Chief executive Dario Amadai and other top US AI leaders have publicly said that most users want the most intelligent AI systems available at any time, suggesting that Chinese systems are less of a threat since their capabilities generally trail those of top AI models by at least a few months. Anthropic is targeting a public debut in September or early October, people familiar with its plans said. Its top rival, OpenAI, is now expected to follow in an IPO that could be as late as next year. In the meetings, Anthropic told some investors that it plans to push further into healthcare and biology AI uses and that the work could help mitigate some of the negative sentiment around AI. People involved in the conversation said Anthropic hasn't shared its precise pricing plans or timing. The people said Anthropic's IPO would come during a banner year for such megadeals, following Elon Musk's SpaceX, which surpassed $2 trillion in trading after its IPO earlier this year. While some bullish analysts and investors expect Anthropic to exceed that, a sell off in SpaceX's shares after early pop has reminded investors of how turbulent moving from private to public markets can be, Anthropic pulled ahead of the pack earlier this year when its Claude code coding tool took off. Anthropic said in May that its run rate revenue had topped $47 billion earlier that month. End quote. Mark Gurman's sources say Apple remains on track to launch a Glass centric iPhone, an overhaul of the iPhone Pro line next year, countering rumors that led Jefferies to downgrade the stock quote. The company expects to launch iPhone Pro models next year that will make use of a new glassy look, according to the people who ask not to be identified. Glass is used on the front and back of the phones known internally as V73 and V74, and the material will curve into the sides of the devices with a metal band in the middle. A more aspirational version championed by Apple's design studio used even more glass instead of metal, but was scrapped early on. With that approach, the company encountered problems connecting the glass panels together. The design didn't hold up when Apple had to figure out how to produce it at large volumes, according to the people. The technology giant has been working to create an eye catching new look for its flagship product, which first debuted in 2007. Though Apple's vision for an iPhone with uninterrupted glass has proven challenging to execute, the company still expects to make a splash next year with an evolution of the original idea. Apple's product roadmap hasn't changed, according to one of the people refuting a report this week from Jeffrey's Edison. Lee, the analyst, had said the all Glass iPhone was canceled, citing supply chain checks. Lee also downgraded Apple shares, saying the move would hinder efforts to offer higher priced smartphones. Apple's shares fell as much as 2.8% in New York on Monday. New iPhone designs are typically solidified around a year before the product's annual debut in the fall. That means the plans for 2027 are already in advanced testing and mostly Locke, barring unforeseen hitches that could still occur before that. A new foldable version will be the biggest single change in the history of the product's look. Apple is also slated to unveil iPhone 18 Pro and Pro Max models with major camera improvements, and it's planning a new dark red color that the company expects to generate some buzz. End quote with this summer's record breaking heat, it's hard to cool down enough to get a good night's sleep. That's where the pod comes in. The Pod by eight Sleep is a smart mattress cover that goes over your existing mattress and actively heats or cools each side of your bed independently. It connects to your wearable, analyzing your daily activity. Then it predicts how you'll sleep and adjust the pod's temperature automatically. What I love about the eight Sleep is the simplest of things the ability to be cool when it's time to go to sleep and warm when it's time to wake up. And my is entirely under my control. My wife can be whatever temperature she wants on her side of the bed too. Use Code ride home@eightsleep.com ride home for up to $350 off. That's code ride home@8sleep.com ridehome. Ever woken up and needed a cup of coffee just to feel like a person? If you're thinking yes daily, then listen up. Ultra Pouches partnered with leading neuroscientists to design these pouches. They use clinically proven nootropics and adaptogens to help deliver immediate focus and smooth energy that lasts one to two hours with zero nicotine and zero caffeine. Yep, it's possible to feel energized and focused with zero nicotine or caffeine. So don't sleep on Ultra Pouches. Literally. New customers can use code TBRH to get 15% off@takeultra.com that's takeultra.com for 15% off with code TBRH. Your favorite YouTube stars are gonna have to work harder to make money. Or I guess they're going to have to work harder to get you to watch them so they can make money. Quoting TechCrunch creators who want to start earning on the platform will need at least 8,000 qualified watch hours over the past year, or 20 million qualified shorts views in the last 90 days. Currently, creators need 1,000 subscribers and 4,000 watch hours over the past year,or 1,000 subscribers and 10 million shorts views over the 90 days. The change is going into effect starting February 1st. The Google owned company says the update won't impact creators already in the YouTube Partner Program. Additionally, YouTube announced that creators will need to maintain 10 million Shorts views over a 90 day period to earn money through the Shorts creators pool. Channels below this threshold will remain in the partner program and continue earning on long form content, with Shorts revenue resuming once they cross the 10 million views barrier again. YouTube says the changes are being introduced to keep pace with the growth of YouTube, which now sees over 200 billion daily shorts views and over a billion hours of watch time on TV every day. The changes make it harder for creators to enter and remain in YouTube's monetization program, especially for Shorts creators. By increasing thresholds, YouTube is putting more pressure on creators ability to consistently bring in large audiences before being able to earn money on the platform, which could in turn lead to fewer new entrants being able to monetize their content creators. As part Of Monday's announcement, YouTube announced that it's expanding its more affordable Premium Lite subscription to all countries where YouTube Premium is available. Creators receive a share of subscription revenue based on member watch time and views, with 55% going to long form video creators and 45% to shorts creators. With these additional subscribers, creators can expect higher earnings When a user signs up for Premium, partners on average earn more than when the user was watching ads, the company wrote in the blog post. Premium Lite offers an ad free experience on most videos, along with the ability to download videos for offline viewing and play videos in the background. YouTube isn't the only social media company revising its creator rewards program over the weekend. Elon Musk's X also revamped creator payouts by changing its guidelines to only reward original content, end quote. One thing I didn't mention yesterday when we talked about Zuck's new AI manifesto was that I guess open weight AI is back on the menu. Quoting Spyglass I'm old enough to remember the last time Mark Zuckerberg was embracing Open as the future of AI. Which is to say, I'm at least 2 years old a lot has changed since the last time Meta released an open weight model, namely, seemingly the entire team building models for Meta. Famously, Llama was put out to pasture as Alexander Wang and his scale team was brought in to fix what ailed those previous models. That included not only spending billions on fresh AI talent and data centers, but also setting aside that notion of open in favor of a closed approach to AI development. Why? Because it sure seemed like that model and those models had won the day just over a year ago. To hear Meta tell it, as I'm sure we're about to hear endlessly, the company never left Open behind. They simply did what they needed to do in order to reset their efforts and catch up fast. And there's undoubtedly some truth in that, but also some gaslighting. If OpenAI and Anthropic were still running away with the AI game, or if Meta was in the lead, we probably wouldn't be hearing too much about Open right now. But again, a lot has changed in the past year. The real key was China seemingly catching up to that frontier of AI. Not quite, but perhaps close enough. Closer than deep seek 18 months ago, and with AI in a more mature state, being close enough perhaps matters more now. And that's especially true given the cost and consumption concerns that a lot of corporate America is feeling at the moment with regard to AI. If those Chinese AI models are close enough at a fraction of the cost of the Frontier models. Meanwhile, the US Open model race still feels wide open, and without anthropic OpenAI or even Google fully committing to it again at the Frontier, Meta may have an opening here. The real key will be what happens when Meta releases their actual Frontier model, the one codenamed Waterman, which many expect to be both coming soon and also competitive with said Frontier. Will that be opened as well at some point shortly after launch? Meta isn't saying. My guess would be that if Watermelon really is as good as the current Frontier, the weights won't be released, but it will instead be used to distill some other Muse variant that Meta will release as an open model. We'll see. Regardless, it's clear that Meta and Zuck in particular are slamming on the gas. That's why you get a 7,000 word essay just a couple weeks after writing a decidedly more parsable version as an op ed in the Wall Street Journal. This memo is verbatim to that at many points. So this is like the Director's Cut, I guess. Boy, it needs an editor. Meta clearly sees Open as a new opening here which yes, is ironic given that it was their initial path. But they also see openings in consumer and personal AI, especially with OpenAI seemingly taking their eye off that ball in their fight with Anthropic, as well as an opening on the business model front where margins are still always opportunities. Still unclear though. Can Meta make money on any of this and have it pay off over time? They seemingly have more options now, but this is all completely unproven. Zuck also clearly sees an opening around the messaging of AI. He wants Meta to hold the positive AI position. That's going to be easier said than done, quite literally. But yes, open is a part of that messaging. So Open is back. No llamas this time though, sadly. Finally today, flight tracking platform FlightAware is suing Kalshi in New York, alleging Kalshi is using its data without permission to let users bet on flight cancellations, quoting the Journal Kalshi has built flight cancellation betting pages verified by FlightAware's data despite demands to stop, according to a complaint filed in U.S. district Court on Monday. According to the suit, CalSHI uses FlightAware's data to settle bets on the percentage of scheduled flights that would be canceled out of an individual airport in a particular time period and therefore whether those wagers will win or lose. FlightAware took issue with the use of its registered trademark on Kalshi's betting pages as concerns mount that running such markets could encourage unsafe tactics to cause cancellations and create the potential for massive disruption of air travel, according to the suit. Lawyers for FlightAware cited instances of alleged betting market manipulation to justify their concerns about gamblers influencing air traffic for profit, including suspected tampering with a weather station at Charles de Gaulle Airport in Paris that turned a polymarket wager of less than $120 into more than $21,000. FlightAware claimed in the suit that it had sent Kalshi cease and desist letters to no avail, and wants Kalshi to stop offering bets on cancelled flights before there is any harm to public safety. Nothing more for you today. Talk to you tomorrow.