 |
|
|
|||||
Its rare that your esoteric, impossible-to-pronounce, decade-long research project becomes a technology so crucial to national security that the President of the United States calls it out from the White House. But thats exactly what happened to Dr. Eric Wengrowski, the CEO of Steg AI. Wengrowski spent nearly a decade of his life advancing steganography, a deeply-technical method for tracking images as they travel through the machinery of the modern Internet, as the focus of his PHD at Rutgers University. After earning his degree, Wengrowki and a team of co-founders rolled his tech into a small startup. For several years, the company grew, but mostly toiled away in relative obscurity. Then, AI image generators exploded into the publics consciousness. And for Wengrowki and Stegs team, everything blew up. Durable Marks I met Wengrowski during the pandemic, when we both volunteered to help a media industry trade group rapidly pivot its yearly in-person conference to a Zoom format. For years, I only knew Wengrowski as a cheerful, highly-intelligent floating head in my video chat window. I even interviewed him for my YouTube channel from the COVID-safe confines of our respective home offices. When I finally met him in person in San Francisco in 2023, I discovered that hes actually a towering 6 feet 3 inches tall. It was one of those iconic pandemic professional meet-cute moments people joke about, where you find that someone youve virtually known for years looks totally different in person. What wasnt different about Wengrowski in real life was his intense interest and passion for his chosen field. Steganography (pronounced STEG-an-ography, like the Steg in stegasaurus) is a technique for embedding an invisible code into the pixels of an image. Basically, a complex algorithm subtly changes selected pixels in a way thats invisible to human perception. Images look no different after being marked with a steganographic watermark than they did before. Yet, when special software looks at the marked image, the unique code embedded in its pixels comes through clearly to the softwares computerized eyes. The presence of that code lets companies like Steg track a marked image back to its source with extremely high accuracy. Crucially, because the code is embedded directly into the images pixels, its also nearly impossible to remove. Bad actors can easily crop out a physical watermark from an images pixels, or use a tool like Photoshop to scrub data from the images IPTC or EXIF metadata fields. In contrast, because steganographic watermarks live directly in the visual part of the image itself, they travel with the image no matter where it goes. And they survive the most common image-related funny business that nefarious people might try to use to remove them. All steganographic watermarks can survive things that amateur image thieves might try, like aggressive cropping, or even the common practice of taking a screenshot of an image in order to stealthily steal it. But Stegs tech goes even further, Wengrowski told me in an interview. If for example you load an image watermarked by the companys tech on your computer screen, take out your phone, and photograph the physical screen, the companys watermarks will survive in the new image on your phone. Your nefarious copy will remain traceable to the original with Stegs tech. AI Explodes Everything When Wengrowski originally developed Stegs technology, he knew it was cool. And he had a hunch that it was useful for something. But exactly what that something might involve wasnt originally clear. In the early days, Steg slowly grew by helping companies with legal compliance and image protection. Steg would embed its watermarks in copyrighted images, for example, and then trace where those images ended up. If someone stole and used a copyrighted image without permission, Stegs embedded watermarks could be used to prove the theft and could help lead to a legal settlement. The company also worked to safeguard things like pre-release images of a new product. If a company sent top-secret images of a new phone (marked with Stegs tech) to a supplier, for example, and those photos suddenly ended up as a leak in TechCrunch, the company could trace the embedded watermark and know who to blame. That was enough for Steg to grow slowly and steadily improve its tech. Then, in 2022, everything changed. All at once, OpenAI released its Davinci image generation model (remember the avocado chairs?), Midjourney rolled out its then world-beating image generation tech, and Google leaned into image generation within its Bard and later Gemini AI models. Almost overnight, the world was awash in AI images. And very quickly, they became so realistic that everyday people had trouble knowing what was real and what was AI generated. This presented a huge problem for AI companies. They wanted to release their tech far and wide. But they fretted about the potential societal (and legal) consequences if their images were used for deepfakes to deceive people, or even to sway elections. And more broadly, anyone with an interest in the veracity of images suddenly had a huge problem knowing what was real and what was AI-generated. Everything from news reporting to war crimes tribunals rely on imagery as evidence. What happens when that imagery can be quickly and cheaply spun up by an AI algorithm? Yes, AI companies can physically watermark their images (such as by adding a little Gemini star in the lower right), or embed Generated by AI markers in their images metadata. But again, removing those markers is childs play for even the least sophisticated scammers. With AI image generators storming the world, the origins and veracity of every image online was suddenly called into question. Thank You, Mr. President That led to a bizarre situation for any deeply technical person pursuing their random, highly-specific passion in relative obscurity. On October 30, 2023, Wengrowki woke to find that then-president Joe Biden had issued an executive order specifically calling out AI watermarking tech, highlighting it as a crucial factor in national security, and ordering all Federal agencies to use it. Specifically, Bidens order mandated embedding information that is dificult to remove, into outputs created by AI including into outputs such as photos, videos, audio clips, or text for the purposes of verifying the authenticity of the output or the identity or characteristics of its provenance, modifications, or conveyance. The order also specifically called for the rapid development of science-back standards and techniques forlabeling synthetic content, such as using watermarking. Biden framed this as mission criticalthe term national security appears 36 times in his executive order. Basically, Biden was mandating the use of tech like steganography, and specifically calling it out from the White House. When that happened, Wengrowski told me, everything went crazy. Since the orderand the corresponding growth of AI imagery more broadlyStegs revenue has increased 500%. Moreover, protecting the integrity of images appears bipartisanWengrowski told me that AI watermarking has been embraced by both the Biden and Trump administrations. In an extremely tight AI job market where top researchers can command eight-figure salaries, Steg now employs five machine learning PHDs devoted to improving its technologies. Although Wengrowski couldnt share his customer list on the record, I can vouch for the fact that its wildly impressive. While keeping its legal compliance and image tracing side alive, Steg has expanded aggressively into the world of cybersecurity and AI image watermarking. For AI companies that want to ply their trade without ruining humanitys trust in visual media, Stegs tech is a lifeline. Companies can embed a steganographic watermark directly into AI images the moment theyre generated. For the life of an image, the code travels with it, even if its reposted, edited or altered. If that image is used as a deepfake or used to manipulate or harass people, the company that created it can quickly read the embedded steganographic watermark in its pixels, definitively label it as a fake, and quickly dispel any damage the image might cause. If youve created an AI image in the last year, youve almost certainly used steganography without even knowing it. Most major AI image generation companies now use the tech. Many use Stegs. And in a world where AI images are so good that they easily fool most detectors (and even trained forensic image analysts), many companies see steganography as the only bulwark against AIs total destruction of any truth still left in the visual world. A Wild Ride For Steg and for Wengrowski personally, its been a wild ride. Right as Biden issued his order, Wengrowski became a father, and now juggles the everyday struggles and joys of a young parent with the rigors of such things as constant travel and testifying in state legislatures. The rise of AI imagery has also revealed some counterintuitive challenges. When Steg first launched, Wengrowski told me, he expected that people would yearn for a technology that could prove whether an image was real or fake. In reality, he was surprised by how little people care. Many people are fine with seeing AI generated content, as long as its funny, informative or otherwise engaging. Whether or not its properly labeled as AI matters very little to them. More pointedly, it matters very little to the social media platforms that disseminate the content, too. Again, though, for the companies who create that contentand who face legal and reputational risk if their tech runs awryit matters an awful lot. Wengrowski tells me that Steg is continuing to improve its tech, making its watermarks even harder to beat. The company is also entering the emerging field of poisoning. New software that Wengrowki showed me invisibly alters images in ways that trip up common deepfake algorithms. If someone tries to turn the poisoned image into a deepfake, it comes out garbled and illegible. The tech works both when images are used for training deepfake models, and when a bad actor tries to create a deepfake of a specific person. The idea is that an influencer, for example, could upload poisoned images of themselves to their social media. The images would look normal to human users. But if someone tried to deepfake the influencer, the poisoned images would thwart them. Wengrowki told me hes especially excited to use the tech to help protect young influencers and teens in general, who are often targeted in abhorrent cyberbullying attacks involving explicit deepfakes. More broadly, though, Wengrowskis story is an inspiring one for anyone grinding away on an as-yet unproven technology, convinced of its value but unsure whether the world will ever see their work. Reflecting on Stegs success, Wengrowski acknowledged that Its probably best to start a business with a clear plan and an understanding of product/market fit. But in his words, Theres also something to be said for knowing a technology is cool, continually improving it even if you have no idea where that will lead, and just trusting that eventually it will have some value for the world. In Stegs case, thats indeed been a winning formula.
Category:
E-Commerce
Robinhood is betting that its customers want to trade on absolutely everything. On Tuesday, the popular stock-trading app unveiled a slate of updates to its prediction markets business, aggressively expanding into sports. Now, Robinhood users can trade contracts tied to specific professional football players’ performances, as well as prepackaged combos for individual games. Early next year, customers will be able to combine up to 10 outcomessuch as winners, spreads, and totalsinto a single, custom-built contract. Robinhood’s news site, Sherwood, is also launching a new sports newsletter, Scoreboard. Eventually, Robinhood plans to launch contracts that span not just multiple games, but multiple categoriesfrom sports to climate to politics. “If customers say they’re looking to trade a specific category or specific event, we’re all ears,” Adam Hickerson, Robinhood’s senior director of futures and prediction markets, tells Fast Company. Robinhood’s entrance into prediction markets can be seen as a natural culmination of the trajectory it set into motion years ago. For the uninitiated, prediction markets allow people to trade on real-world events by buying and selling contracts. These events can range from sports matches to political elections to who Time magazine will name as its Person of the Year. Since Robinhood launched prediction markets in late 2024, they’ve become the companys fastest-growing line of business. In the third quarter this year, it reported that users traded 2.3 billion prediction-markets contracts. Then, in October alone, that figure reached 2.5 billion. The most popular contracts have been in sports. Robinhood maintains that users are not gambling, as trading on the market sets the odds, not the platform itself. Still, the sports contracts tap into an enthusiasm for sports speculation. According to the Pew Research Center, 22% of adults in the U.S. have bet money on sports in the past year. Among men younger than 30, that figure rises to 36%. Robinhood isn’t the only app getting in the game: Fanatics, a global sports platform, just launched a prediction-market app, becoming the first sportsbook to do so.Some regulators believe decision markets cross the line into gambling. Numerous states have sent Robinhood cease-and-desist letters, demanding prediction markets stop offering sports contracts. In response, Robinhood sued New Jersey and Nevada earlier this year, maintaining that its markets are completely legal. There is no sign of regulatory turmoil dampening Robinhood’s ambition. In November, the company announced plans to start its own prediction market, launching an exchange with Susquehanna International Group. “This is just the tip of the iceberg,” Hickerson says. From meme stocks to prediction markets In 2013, Vlad Tenev and Baiju Bhatt founded Robinhood as an easy, commission-free way for users to trade on their phones. Business boomed during the COVID-19 pandemic as bored Americans, flush with stimulus checks, flocked to the app. The online brokerage became known for its popularity among young adrenaline junkies who treated investing less like retirement planning and more like a mobile game. Robinhood’s supporters applauded the company for democratizing finance. Critics, meanwhile, slammed it for encouraging users to trade stocks like gamblers betting on sports. In 2021, legendary investor Charlie Munger told CNBC that Robinhood was “a gambling parlor masquerading as a respectable business,” calling it a “sleazy, disreputable operation.” The anti-Robinhood backlash boiled over during the GameStop saga, when users fueled a trading frenzy that drove meme stocks to absurd highs. In the aftermath, Robinhood dialed back on its so-called gamified elements, removing a confetti animation that accompanied certain achievements. In a February 2021 hearing before Congress, Tenev testified that the “vast majority” of Robinhood’s customers were long-term investors, not day traders buying meme stocks. Eventually, however, Robinhood acknowledged the importance of its active day tradersusers less interested in traditional stocks and far more interested in riskier products like cryptocurrencies. These are our most engaged customers that generate the lions share of our revenue, Tenev told The Wall Street Journal in November. We put our best people on active traders. Robinhood keeps these valuable customers happy by letting them invest in whatever their hearts desire. Increasingly, that means prediction markets. Betting on predictions Prediction markets have been around for more than a century. They have primarily existed as a niche curiosity, not a major focus for investors, amateur or professional. Then came last years election. More than $3.3 billion was traded in 2024 presidential election contracts, mostly on prediction market Polymarket. Robinhood launched its own contracts a month before the election, its first foray into the prediction markets. By the time Trump wonas forecast by the marketsthe concept of prediction markets had cemented itself in the American mainstream. At the time, it was not legal for Americans to use Polymarket. As a result, it operated offshore, although Americans likely still used it via VPNs and thanks to Polymarket’s reliance on cryptocurrency. Reports alleged that much of the election-trading volume came from “wash trading,” which inflates market activity and is a form of market manipulation. In January, Kalshi launched “100% legal” sports trading in all 50 states, regulated by the Commodity Futures Trading Commission. The next month, Robinhood announced a partnership with Kalshi for Super Bowl contracts. But mere hours after the announcement, Robinhood canceled the contracts at the request of the CFTC. The agency had serious concerns that the contracts may not be permissible under the law, a CFTC representative said at the time. Robinhood was undeterred. In March, just ahead of the NCAA basketball tournament, the company launched its prediction markets hub, still in partnership with Kalshi. A CFTC official told Sportico that the agency had no legal justification to prevent Robinhood from offering access to these contracts.”Robinhood is competing not only with Polymarket and Kalshi but also with Coinbase, which is reportedly planning to introduce prediction markets, thanks to its own partnership with Kalshi. Defining “gambling” Laws on gamblingand how regulators interpret those lawsare set to be one of Robinhood’s biggest obstacles when it comes to decision markets. Kalshi and Robinhood maintain that their users are trading, not gambling. Kalshi and Robinhood do not make money based on outcomes, but instead earn revenue via transaction fees. They emphasize that, unlike sportsbooks, prediction markets do not take bets or set odds. For example, if the New England Patriots are playing the Tennessee Titans, a user who thinks the Titans will win could buy shares on the “yes” position. Shares will be less expensive if Tennessee is the underdog, but the price is set by the markets’ perceived probability. If the Titans win, each winning share pays out $1, while those who picked the Patriots are left empty-handed. If regulators agree with decision markets’ line of reasoning, companies like Robinhood and Kalshi should be able to offer sports contracts online in all 50 states, including states where sports betting is illegal or restricted. Gambling is banned for people under 21, but 18-year-olds are typically allowed to trade event contracts. Many sports-betting regulationssuch as safeguards to prevent game fixingdo not currently apply to prediction markets. Not every regulator is convinced by decision markets’ arguments. In addition to New Jersey and Nevada, Connecticut, Ohio, Maryland, Illinois, and Arizona have sent Robinhood and other decision markets cease-and-desist letters. Nonetheless, the decision markets have amassed some powerful allies. In January, Donald Trump Jr. joined Kalshi as a strategic adviser. Polymarket’s return to the U.S. and its legalization came, in part, thanks to the president’s son becoming an investor and adviser. In September, President Trump nominated Kalshi board member Brian Quintenz to chair the CFTC. A month later, Trump Media announced a prediction market partnership involving Crypto.com and Truth Social. Everything, everywhere, all at once Folding sports contracts, stock trades, online banking, and crypto into a single app will inevitably upset traditionalists. Old-school investors might also be skeptical of Robinhoods other announcements on Tuesday, focused on artificial intelligence: upgrades to its AI-powered investing assistant and the launch of personalized daily Digests that analyze users’ portfolios. For Robinhood, the grab bag of choices is the point. “Everything comes down to: What does the customer want?” Hickerson says. Robinhood takes feedback seriously, he said, and executives have heard “loud and clear that these are some features that they really want to trade on.” Scrolling through the contracts on any prediction market reveals just how many topics inspire speculation. As of Monday, more than $17,000 in contracts had been traded on Kalshi related to the topic: “What will Vlad Tenev say during the Robinhood keynote?” (Trading activity indicates a 72% chance that Tenev says “sport.”) “Ultimately,” says Oren Naim, Robinhood’s vice president of platforms, “our long-term vision for the company is to become your one-stop shop for anythingany financial needacross the board.”
Category:
E-Commerce
Suzanna’s Kitchen, a Georgia-based food production company, has issued a recall of 62,550 pounds of fully cooked, bone-in breaded chicken products. The chicken, which was distributed nationwide, was recalled over mislabeling. While the product was labeled with a product code that classifies it as non-allergen-containing, the product contains soy. According to the recall notice, which was issued on December 12, the affected product is the eight-piece cut, bone-in breaded chicken portions that were produced on October 16, 2025. The U.S. Department of Agriculture (USDA) mark of inspection and establishment number printed on the side of the package is P-1380. According to the USDA, soy is one of the “big nine” allergens and could result in serious allergic reactions. “Symptoms of food allergies typically appear within minutes or up to two hours after a person has eaten or has come into contact with the food to which they are allergic,” the department’s website explains. It also notes that common signs of an allergic reaction include hives; difficulty breathing; swelling of the tongue, lips, face, throat, and vocal chords; a drop in blood pressure, and more. It’s unlikely that the products will be found in home refrigerators, as it was distributed to restaurants across the country. However, restaurant-goers with soy allergies should be aware of the heightened concern. The USDA’s Food Safety and Inspection Service (FSIS) says restaurants should carefully check their stock. FSIS is concerned that some products may be in restaurant refrigerators or freezers. Restaurants are urged not to serve this product; these items should be thrown away, the recall notice states. The notice also said there have been no confirmed illnesses due to the affected products, but that consumers concerned about a potential illness should contact their healthcare provider immediately. Otherwise, questions about the recall can be directed to Dawn Duncan, Customer Service Director, Suzannas Kitchen, at dduncan@suzannaskitchen.com, the notice states. The USDA’s Meat and Poultry Hotline is also available for questions at 888-674-6854.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||