Bloomberg: Sources: Denver-based, Blue Owl Capital-owned Stack Infrastructure is considering selling all or some of its Asia data center operations in an up to $30B deal  —  Stack Infrastructure Inc., a data center company owned by Blue Owl Capital, is considering options including a sale of its Asia operations …

Activists, a leading labor union, and a big retirement fund are asking challenging questions about what’s expected to be the largest-ever stock market debut.

PLUS: Make your Notion agents more autonomous

Financial Times: DefiLlama: investors pulled ~$14B from the DeFi sector after North Korea-linked hackers stole $290M from Aave in April, weeks after stealing $280M from Drift  —  DeFi was once touted as the future of finance but traders have grown concerned over the security of these projects

Learn how ChatGPT safeguards your privacy, reduces personal data in training, and gives you control over whether your conversations improve AI models.

Todd Feathers / Wired: A medical student reverse-engineered AI tools used by medical colleges on suspicion they were filtering his applications, highlighting AI-driven hiring concerns  —  Armed with some Python and a white-hot sense of injustice, one medical student spent six months trying to figure out whether an algorithm trashed his job application.

QyTw0, the Finnish AI lab founded by former AMD Silo AI CEO Peter Sarlin, is now valued at €325 million (approximately $380 million) after raising a €25 million angel round ($29 million). It's a sign of enduring tailwinds for AI, quantum computing, and sovereign tech, especially for Europe-made companies.

Zelenskyy released new videos of the FP-5 Flamingo launching in combat on Tuesday. Volodymyr Zelenskyy Zelenskyy published new footage on Tuesday of Ukraine's Flamingo missiles taking off in combat. They were used to strike a military-industrial complex deep inside Russia. Footage of the attacks showed large explosions and fires in the Russian city of Cheboksary. Ukraine has published new clips of its homegrown FP-5 Flamingo missiles launching into battle, saying they flew some 930 miles to reach a military manufacturing complex deep inside Russia. The videos, posted by Ukrainian President Volodymyr Zelenskyy on Tuesday, provide some of the clearest footage yet of the missiles taking off from the ground. The clips appear to show the Flamingos igniting their jet engines for an initial burst of acceleration, rising quickly into the sky before gently arcing into a forward trajectory. Their manufacturer, the Ukrainian firm Fire Point, has said that the 40-foot missiles have a maximum range of 1,860 miles and can carry a 1,150-kg warhead at speeds of up to 560 mph. Zelenskyy said in his post that the missiles shown on Tuesday had "covered a distance of more than 1,500 kilometers," making it one of their longest announced combat flights. Zelenskyy added that a major target for the strike operation was a military-industrial complex in Cheboksary, a Russian city about 600 miles from Ukraine's northern border. "The plant supplied navigation elements for the Russian Navy, missile industry, aviation, and armored vehicles," the Ukrainian president wrote. Russian Telegram news channels and military bloggers reported on Tuesday that the VNIIR-Progress complex in Cheboksary had come under Ukrainian attack on Monday. Several videos and photos posted by these channels appeared to show large explosions in the city. One clip showed flames and debris covering the main entrance of the ABS Electro building, an enterprise in the industrial complex that builds satellite guidance systems for Russian weapons, submarines, and drones. Another photo, first geolocated by the Russian Telegram channel ASTRA to a small strip mall about 800 yards away, showed black smoke rising from a large fire in the complex's general direction. Business Insider was unable to verify when these clips or photos were taken. The Institute for the Study of War, a Washington-based think tank, said in a Tuesday assessment that footage on social media indicated six Flamingo missiles had struck the VNIIR-Progress plant. A long-range fixed-wing drone was also filmed crashing into what appears to be a residential building. Russian state media outlet RIA Novosti reported that two people were killed and another 32 were injured in a "massive attack" in the region. The VNIIR-Progress plant previously came under attack from Ukrainian drones in February, June, July, and November 2025, with regional head Oleg Nikolaev saying the June strike temporarily halted production. Locally designed and produced, Ukraine's Flamingo missiles have taken the spotlight in the country for providing Kyiv with a long-range attack option that can threaten Russia's military-industrial base. The West has plenty of such weapons, but has so far refrained from giving them to Ukraine for strikes on Russian soil. Apart from the Flamingo, Ukraine has more commonly relied on slower one-way attack drones, such as the fixed-wing Liutyi, to hit facilities deeper inside Russia. While cheaper and more numerous, they're easier for air defenses to intercept. Zelenskyy said last year that Ukraine was planning to manufacture seven Flamingos per day. Read the original article on Business Insider

Financial Times: Sources: DeepSeek is in talks to raise funds, and the Big Fund, China's biggest state-backed chip fund, is seeking to lead the investment at a ~$45B valuation  —  Value soars in ongoing fundraising discussions as investors including Tencent seek slice of AI lab

Emil Ryd / @emilaryd: Study: using weaker AI models to supervise a more capable model could prevent the stronger model from deliberately underperforming on benchmarks and evaluations  —  New paper from MATS, Redwood, and Anthropic! If a capable model is strategically sandbagging, can we train it to stop when the only supervision we have comes from weaker models? We find that we can! Work done as part of the Anthropic-Redwood MATS stream. [image]

Wonder wants to turn its robotic kitchens into AI-powered “restaurant factories,” letting anyone spin up a virtual food brand with a prompt.

Financial Times: Accounting software company Visma's delayed IPO, coming amid “SaaSpocalypse” fears, is a major setback for private equity parent Hg, a software buyout giant  —  Delayed IPO of €19bn software group Visma was meant to be a crowning moment for PE group Hg — and London

Christina Kyriasoglou / Bloomberg: Infineon expects data center revenue to grow from €1.5B in FY2026 to €2.5B in FY2027, as it and EU peers STMicro and NXP benefit from AI infrastructure demand  —  Infineon Technologies AG forecast revenue that beat analysts' expectations in the current quarter as the German chipmaker benefits …

Look inside OpenAI's GPT5.5 party. Olivia Moore on X OpenAI let ChatGPT plan a launch party to celebrate the release of its latest model, 5.5. ChatGPT wanted the event to be held on May 5 at 5:55 pm, with its human creators delivering a toast. Pics from attendees show an Apple Store-like experience. OpenAI's invitation to GPT-5.5's launch party said the vibe would be intimate: a "low-key" gathering with CEO Sam Altman, drinks, swag, and a few hundred developers celebrating the company's latest AI model at its San Francisco headquarters. Instead, it became one of the hottest tickets in tech, with more than 8,000 people scrambling to get into a party that Altman says ChatGPT itself helped plan. While it remains unclear how many applicants were ultimately offered access to the exclusive event, those who were accepted were quick to tout their status on social media. However, the rejects still got a party favor: OpenAI offered upgraded Codex rate limits to everyone who requested a ticket, granting additional free access to its coding agent. Inside, photos posted by attendees show the vibe was somewhere between a developer meetup, a Hollywood premiere, and a very expensive social experiment. OpenAI did not respond to a request for comment from Business Insider. An invite felt like a golden ticket Is today!! see you in the @OpenAI HQ GPT-5.5 Party pic.twitter.com/8xGPUnIngQ — Diego | AI 🚀 - e/acc (@diegocabezas01) May 5, 2026 It looked like an Apple store opening At the OpenAI 5.5 party now. pic.twitter.com/Tt0UvycxXo — Ahmad Awais (@MrAhmadAwais) May 6, 2026 Everyone wanted a photo with sama really dope party 🔥🔥🔥@sama @OpenAI pic.twitter.com/3crw3ln6kH — alpha_rover (@Alpha10six) May 6, 2026 Met a really cool person at an AI party😊 @openai @sama pic.twitter.com/ODouAKRYLm — Derya Unutmaz, MD (@DeryaTR_) May 6, 2026 People wore goblin ears went full goblin mode with @sama https://t.co/yUg0rO0PnS pic.twitter.com/AOT2nNvbgN — Kate Deyneka (@katedeyneka) May 6, 2026 Swag included nerdy socks a few things i collected from the party pic.twitter.com/9LOVnwNL7P — nic (@nicdunz) May 6, 2026 Someone bought a Sama Lisa? Such an awesome party, very cool talking to you @sama and the Codex team! pic.twitter.com/bvca8bQqJJ — thegraxisreal (@thegraxisreal) May 6, 2026 And he was asked to sign it update - they have brought him a sama lisa pic.twitter.com/KAvlIPh8nr — Alex (@AlexanderTw33ts) May 6, 2026 Read the original article on Business Insider

Discussion | Link

Scott Calvert / Wall Street Journal: Security camera companies are embracing AI to give customers detailed descriptions of surveillance footage that are often spot on but can also be wildly wrong  —  AI descriptions, designed to give customers more detail than generic motion alerts, are mistaking brake lights for house fires and humans for bears

As the war with Iran continues, breach attempts targeting the United Arab Emirates tripled in a few weeks — many targeting critical infrastructure.

Match Group's CEO said Gen Z are intimidated by traditional dating apps. Nordin Catic/Getty Images for The Cambridge Union The CEO of Hinge and Tinder said traditional dating apps intimidate Gen Zers. He said they are "highly structured," and Gen Z don't want dating to "feel like a job interview." Match Group is trying to get users to meet in real life, through physical dating events. The man behind Hinge and Tinder said Gen Z doesn't want dating to feel like "a job interview." In a Tuesday earnings call, an investor asked Match Group CEO Spencer Rascoff about the rise of offline dating activities, such as run clubs and book clubs, and how they affect online dating. The dating apps executive answered that Gen Z "desperately wants to connect" and meet new people. "They just want to do it in a low-pressure, low-stakes way that doesn't feel like a job interview," Rascoff said. He added that traditional dating apps are "highly structured and can be intimidating to a user under 30." The rise of these offline dating events shows that Gen Z is seeking easy ways to find their significant others, he said. To align its apps with the offline dating trend, Match Group has introduced several features that encourage users to go on dates or meet in person. It is rolling out a "Direct to Date" function on Hinge that prompts users to go on dates immediately after matching, cutting out small talk. Rascoff said that in the testing stage of "Direct to Date," users are "defaulting to familiar low effort date ideas like dinner, drinks, and walks." And in March, Tinder launched dating events where users could meet in person, starting with meet-ups in Los Angeles. It's also trying to help users connect organically, with new features like "music mode" and "astrology mode," to get users to share similar interests. "We are embracing this trend of meeting people IRL at different, in different modalities rather than hiding from it," Rascoff said. Match Group has a large portfolio of dating apps, including Tinder, Hinge, and OkCupid. The group reported first-quarter revenue of $864 million, up 4% compared to the same period last year. Its stock price is up about 24% in the past year. However, its biggest player, Tinder, is seeing its user base shrink. Rascoff said on the earnings call that Tinder's monthly active users in March were down 7% compared to the same period last year. Traditional dating apps are losing favor with younger users, who are experiencing swipe fatigue. Read the original article on Business Insider

Maria Laposata and her husband took a year off to travel; she found two-week vacations difficult when they got back. Provided by Maria Laposata After a yearlong trip with her husband, Maria Laposata realized that she hated two-week vacations. She found that returning from shorter trips left her feeling worse than when she left. She and her husband opted for a monthlong working trip instead — their first was to France. For the past three years, I have nursed a growing hatred for the standard two-week vacation. In 2023, I took an adult gap year with my husband, Nick. After a year filled with safaris in East Africa, learning to haggle in Morocco, and island hopping in Southeast Asia, we returned home to LA and feverishly rebuilt our lives. I made reluctant peace with the fact that our future travel ambitions would have to fit into neat two-week sprints by obsessively planning. Our first vacation after coming home The itinerary was simple. We'd start by visiting two Greek islands: Syros, where we'd join a multi-day cooking school, and Paros, where we'd lounge around on the beach. Then we'd fly to Malta to explore historical sites before closing out the trip with a weekend layover in Dublin, one of our favorite places in the world. At first, the trip seemed like a success. Our hard-earned travel skills were well-utilized. We breezed through the airport with one backpack each. Syros, our first stop, was pleasantly calm — filled with pottery stores I loved to peruse. The cooking school was intimate and involved — ideal for experienced cooks like us. But when cooking school was over, a few days into our trip, the two-week clock began to nag at me, instilling pressure to make the most of any free time. Each subsequent activity that didn't fill my soul felt like an incrementally larger catastrophe. Snorkeling in barren, over-fished waters was a waste of an afternoon. Beach days marred by clouds and chilly water were a disaster. And eating at a slew of disappointing restaurants in Malta was the tragedy of the century. Laposata and her husband in Morocco during their adult gap year. Provided by Maria Laposata Returning home after vacation When our plane home touched down in LA, I was filled with dread. The thought of opening my work laptop the next day was already taunting me. Worse, I realized that feeling of dread was familiar. I'd felt this way coming home from virtually every vacation I'd taken before. I'd always accepted it as a given. I'd accepted that a vacation could leave me feeling worse than when I'd left. But what was the point of that? There had to be something we could do to change this dynamic. This became a recurring late-night conversation between my husband and me. "Setting aside our adult gap year, did we ever take a trip where we came back happier than we left?" I asked him one night. "Yeah, Alaska," Nick said without hesitation. The answer didn't surprise me; he's an outdoorsman. "But that wasn't really a vacation." Years earlier, we'd spent a month in Alaska during the summer. We were working remotely, but it didn't feel like work. We used a long weekend to explore huge stretches of the state — camping beside a glacier on the Kenai Peninsula with no one else around, watching grizzly bears in Denali National Park, and listening to audiobooks together on long drives back to Anchorage. A few years back, they had spent a summer in Alaska. Provided by Maria Laposata Working during the trip wasn't always convenient, but the slight time difference gave me more uninterrupted time to focus. And with the sun setting so late, our evenings felt expansive. We hiked local trails, tried many must-visit restaurants, and wandered through art galleries around Anchorage. Having something to look forward to each night made it easier for me to log off at a reasonable hour — something I struggled with at home. Best of all, staying plugged into daily life meant I didn't dread returning. That's when it hit me: We should do it again. A monthlong working holiday instead I pitched the idea to Nick — skip a typical two-week vacation so that we could explore more deeply without the pressure to maximize every moment. We'd still take time off, but spread it out. He was in. In January of this year, we put the idea to the test and spent a month in the South of France. Nick got the arrangement approved at work without issue. I had since launched my own boutique travel company, Travelries, and dutifully approved my own working arrangement. Laposata and her husband fell in love with Nice while working in France. Provided by Maria Laposata While working, we fell into a rhythm In Nice, we woke up to the smell of fresh baguettes, shopped for groceries at picturesque local markets, and once a week picked a new seaside town in the French Riviera to visit for dinner. On weekends, we took the train to other regions — bopping between villages in Provence, Monaco, and the Italian Riviera, admiring the glittering coastal views along the way. We never felt rushed, but by the end of the month, we'd visited more than 20 towns. We wrapped the trip with a long weekend in Paris, where I'd once studied abroad and now got to play tourist. This time, the trip home didn't come with the usual dread. We were sad to leave Nice, but we felt full. And instead of bracing for reality, I was already thinking about where we'd go next. Read the original article on Business Insider

arXiv:2605.03092v1 Announce Type: new Abstract: While sentiment analysis is the staple of financial NLP, capturing the nuances of 'why' behind that sentiment remains a challenge. There have been attempts to address this by analysing investor emotions alongside sentiment; however, this does not provide the additional granularity required to understand the target of the emotion/sentiment. We address this by augmenting the StockEmotions dataset with semantically structured opinion graphs, which provide granular semantic depth to the existing sentiment and emotion labels. Using a declarative LLM pipeline, we augment the StockEmotions dataset with opinion graphs for each sentence, derived from 10,000 comments collected from StockTwits. In addition, we study the effect of introducing opinion semantics on baseline classifiers using Graph Neural Networks (GNNs). Our analysis demonstrates that incorporating opinion semantics improves classification performance across different emotional spectrums