Bill Gurley, the Silicon Valley venture capitalist and a general partner at Benchmark Capital, said he has concerns about Anthropic's plans. Michael Kovac/Getty Images for Vanity Fair Tech investor Bill Gurley said he is worried about Anthropic's ultimate goals. Gurley said that Anthropic wants to build "a species that's superior to humans." He said this God-like pursuit is more troubling to him than Anthropic trying to squeeze out competitors. Venture capitalist Bill Gurley says he's come up with a scarier theory than the idea that Anthropic wants to stifle its competition. "The more I dig, I've met people who I dare say think it's their responsibility and they're excited about building a species that's superior to humans," Gurley said during a recent episode of the "All-In Podcast." Gurley, who said he spent a month reading everything he could about Anthropic, said he came away convinced that one of the leading AI startups wants to create a God-like system to rule humanity. "I don't think they think they're writing software. I think they're midwifing a deity here," Gurley said. "And I don't know which one I'm more afraid of, the regulatory capture or the second theory, I call the Dr. Frankenstein theory." Gurley said his "Dr. Frankenstein theory" is "more scary to me" than the often-repeated criticism that Anthropic's push for AI safety requirements is a veiled strategy to lock out its competitors. One of the impetuses for this theory, Gurley said, was Amodei's October 2024 essay titled "Machines of Loving Grace," which takes its name from a 1960s poem by Richard Brautigan that imagines a world in which intelligent machines watch over a humanity freed from the tasks of daily life. In the essay, Amodei hypothesized several ways the economy will change once "powerful AI" is achieved, ranging from a universal basic income to a "capitalist economy of AI systems." "He says, 'It could be a capitalist economy of AI systems which then give out resources to humans based on some secondary economy of what the AI systems think makes sense to reward in humans.' So that's envisioning a deity of sorts that's going to break ties and decide what humans do," Gurley said, quoting from the essay. Gurley's theory is based on his views of how Amodei and other key Anthropic officials have described artificial general intelligence (AGI) or a similar theoretical moment when AI surpasses human capabilities. The exact definition of AGI is hotly debated. Anthropic has not explicitly said it plans to build an AI deity like the one Gurley described. A representative for Anthropic told Business Insider it did not have a comment at this time. Anthropic cofounder Chris Olah, whom Gurley referenced, recently joined Pope Leo XIV at the presentation of Leo's encyclical, a formal declaration of Church Doctrine that focused on AI and humanity. Leo warned against those who seek a transhuman or posthuman existence, extolling the virtues of being human. "The machinery that makes this possible is the work of math and programming and science," Olah said at a presentation of the encyclical in Vatican City. "But what character we choose, how it interacts with the world, how it ought to interact with the world — these are more clearly questions for the humanities, for religion, for philosophy, for society at large." Gurley remains at Benchmark Capital, though he scaled back his ties to the firm in 2020. He publicly defended Benchmark's $75 million investment in Manus, an AI startup founded in China, which Meta later acquired for $2 billion in December 2025. The cohosts of "The All-In Podcast," some of whom have publicly clashed with Anthropic, echoed Gurley's concerns about Anthropic's ultimate aims. "They believe that they're so powerful, these individuals, that they can create God and that by creating God, they are like this Prometheus kind of species," cohost Jason Calacanis said. David Sacks, who served as President Donald Trump's AI and crypto czar, said that it's easy to draw a line from Anthropic pursuing a powerful enough AI that requires safety regulations to advocating an agenda that amounts to regulatory capture. "If you brand yourself as the safe AI company and then try to characterize everybody else as a reckless player and reckless AI needs to be stopped, you can see how this would basically further your monopolistic control over this industry," Sacks said. In the past, Anthropic CEO Dario Amodei has said that he has pushed for stronger regulations because he is "deeply uncomfortable" about AI companies policing themselves since that would leave enormous power in the hands of a few individuals. "This is one reason why I've always advocated for responsible and thoughtful regulation of the technology," Amodei told "60 Minutes" in November 2025. Amodei has also said that if people doubt Anthropic's sincerity, they should look to its past when safety concerns led the startup to hold back on the public release of Claude before OpenAI released ChatGPT in November 2022. "We probably ceded the lead on consumer AI because of that," Amodei told billionaire and investor Nikhil Kamath during an interview on Kamath's podcast in February. More recently, Anthropic said it held back on the public release of Claude Mythos due to concerns that it could expose widespread cybersecurity vulnerabilities. Last week, Anthropic said Mythos-level models will be available to customers "in the coming weeks." On Monday, Anthropic announced that it had confidentially filed an S-1 draft with the U.S. Securities and Exchange Commission, a major step toward what could become the largest initial public offering in history. Read the original article on Business Insider

It's Microsoft's least-weird attempt at a high-end mobile workstation.

Chef Samuel Kim shared what dishes he always orders at Korean barbecue. Courtesy of Baekjeong Chef Samuel Kim shared the dishes you should always order at Korean barbecue and what to skip.  Don't only order beef. Kim loves to add pork belly or pork jowl.  Stay away from the grilled fish and the udon noodles.  With a plethora of meats, appetizers, and sides to sample, trying to narrow down your order is never an easy task at a Korean barbecue restaurant. That's why we asked chef Samuel Kim of Baekjeong to help. Kim expanded Baekjeong, which currently has eight locations across California and Washington, to bring Korean barbecue to more Americans. And after revealing the red flags to look out for at a KBBQ restaurant, he was happy to share the dishes he always orders — and which ones to skip. When it comes to meat, don't only order beef Baekjeong is a popular Korean barbecue chain. Courtesy of Baekjeong Kim always orders beef at Korean barbecue — he's partial to brisket and short rib — but the chef also makes sure to get pork. "Some people will just go for beef, but pork is such an important animal to grill for us at Korean barbecue," he told Business Insider. "Especially pork belly. Always the pork belly." Kim said he also orders pork jowl whenever he spots it on the menu. "It lends really well to the hot quick grill, and it's just a delicious cut of meat," he added. Make sure to include Korean pancakes and stew The kimchi stew at Baekjeong. Courtesy of Baekjeong From calamari to kimchi, every Korean barbecue place puts its own spin on the classic Korean pancake dish. "Everyone has their own recipe," Kim said. "Korean food is a lot like Italian food, where everyone's mom thinks they make it the best — and that's how the pancakes are as well." Kim said it's also important to have some kind of soup or stew (known as jjigae) when you're having Korean barbecue to help "push the flavor of the protein." He recommends trying one with soybean or kimchi. If you drink alcohol, soju and beer are an essential part of KBBQ Soju is the national drink of Korea. 4kodiak/Getty Images/iStockphoto The grain-based spirit soju is the national drink of Korea. Kim said he always drinks some when enjoying Korean barbecue. "It's a key part of the whole experience," Kim said. "I get some people don't drink alcohol, but if you do, give it a shot." Just remember to follow the Korean custom and never pour your own drink — it's considered bad luck! Stay away from grilled fish While Kim loves a seafood stew or pancake with his Korean barbecue, the chef told Business Insider he'd never order grilled or braised fish. "When we decide to go out to eat in Korea, we don't decide on the restaurant we want to eat at — we decide the dish we want to eat," Kim explained. "So you figure out that dish, then you go to the restaurant that specializes in that dish." "If you're going to a Korean barbecue restaurant to eat meat, I don't know why you'd eat seafood," he added. "Personally, I never order any fish." And skip the udon noodles "Sometimes, you'll see Korean barbecue restaurants that try to be a jack-of-all-trades, and their menu is 30 pages long," Kim said. "So, if you ever see udon noodles or anything like that at a Korean barbecue restaurant, I'd stay away." Read the original article on Business Insider

Michelle Ma / Bloomberg: Gigascale Capital, a climate tech VC firm co-founded by former Meta CTO Mike Schroepfer, closed a $250M fund to back early-stage startups supporting the AI boom  —  Gigascale Capital closed a $250 million fund for early-stage startups.  —  Climate tech venture capital firm Gigascale Capital …

Discussion | Link

Anyone who has downloaded affected Red Hat packages should investigate immediately.

Nvidia is combining Unitree’s humanoid hardware with its own AI and simulation tools, in a new design aimed at researchers and developers.

Even Danish researchers think it's bizarre.

Job-hopping could help early-career engineers find their value, a top OpenAI researcher wrote on X. Tara Moore/Getty Images An OpenAI researcher wrote an X thread defending "job-hopping." He said it could help early-career engineers find a fit and know their value before committing to a company. Still, it's a tough time to job-hop, especially in the tech sector, recent data suggests. Gabriel Petersson said he doesn't think young workers need to avoid the dreaded "job-hopper" label. In several X posts on Sunday, the OpenAI researcher said young tech workers should test out different teams before anchoring themselves to one company. His view: early-career engineers need data points — on research projects, culture, and their own market value — before making a long-term bet. He called the traditional stay-put advice "braindead." "Please don't take the advice that you should stay at a company long and 'not hop around' for your first jobs," he wrote. please don't take the advice that you should stay at a company long and "not hop around" for your first jobs it's absolutely braindead to decide on a long term bet with zero datapoints on what a good team looks like and long before you have priced yourself into the market — gabriel (@gabriel1) May 31, 2026 Recruiters have traditionally warned against job-hopping — or, relocating to new positions and companies every one to three years — because it can call a worker's commitment into question. Petersson said he doesn't agree with that advice, suggesting that other young engineers instead take a fast approach to building their careers. "Just tell people you are looking for internship or you want to 'try working together for a month' or say you're a contractor," he wrote, saying that hopping can result in "huge wins for everyone that all sides have information so you can price yourself in." Petersson himself hopped around before landing an AI research position at OpenAI in 2024, when he was 23, according to his GitHub profile. Before that, he worked as a software engineer at Dataland and Midourney for less than two years each, according to his LinkedIn profile. He dropped out of high school in Sweden at 17 to focus on building AI start-ups. He isn't the only tech personality to make the case for job-hopping. In April, when asked to grade some common career advice, Ryan Roslansky, LinkedIn's former CEO, gave job-hopping for more money an "A." A tough job market raised the stakes for young workers Still, the suggestions land at a tough moment for young workers, especially in tech. Tech companies, including Meta, Oracle, Microsoft, and Block, have announced major layoffs in recent months. Challenger, Gray & Christmas, a layoff-tracking firm, said that while fewer employers are cutting jobs overall this year, layoffs in the tech sector are up 40%. Entry-level and engineering jobs have also been hit hard by AI, making the post-college job hunt harder for many young Americans — and raising the stakes for those who do land a role. But Petersson said some of the best engineers he knows spent years at early jobs that, in hindsight, were not valuable stepping stones. He gave the example of an engineer spending "2.5 years at a startup" after college, making $80,000, before later landing a multimillion-dollar deal. He said that some workers do strike it rich by joining the right company early. He wrote that a small number of people end up at a frontier AI lab or as early engineers at fast-growing startups, making life-changing money. "These are extremely rare," he wrote. "Ask any great engineer and you'll realize how many years they wasted with bad companies." Petersson and OpenAI did not immediately respond to requests for comment. Read the original article on Business Insider

After a disgruntled security researcher published several zero-day exploits in recent weeks, Microsoft seemingly indicated criminal charges were in order.

Altman has an "utter disregard" for human lives, Florida AG says.

Enterprise Document Intelligence [Vol.1 #3] - Why the ML toolkit (hyperparameter sweeps, train/test splits, explainability frameworks) solves the wrong problem, and what to use instead The post RAG Is Not Machine Learning, and the ML Toolkit Solves the Wrong Problem appeared first on Towards Data Science.

Discussion | Link

Corgi cofounder Nico Laqua (not pictured) said he has a mattress in the startup's office and often showers at Equinox. Maddy Grassy/Getty Images for ONIT Corgi cofounder Nico Laqua said he sleeps in the office, often gets 3-4 hours of rest, and doesn't take weekends off. "I would rather measure my lifespan in victories than years," he said on the "20VC" podcast. Laqua is one of many AI founders flaunting their hardcore work schedules. Nico Laqua said he'd rather shave years off his life than see his startup fail. He cofounded Corgi, an AI insurance company, in 2024. Corgi became a unicorn in May, raising its Series B at a $1.3 billion valuation — before raising a Series B1 round at a $2.6 billion valuation three weeks later. To reach such heights, Laqua said he keeps an unusually strict schedule, where he works seven days a week, sleeps in the office, and gets around three hours of shut-eye. On an episode of the "20VC" podcast released Saturday, host Harry Stebbings asked Laqua: "Would you rather Corgi was a trillion-dollar company, but you died at 50, or it was a fail, and you live till you were 80?" "The answer to that is pretty easy," Laqua said. "I'm dying either way." Laqua's outlook is part of a bigger trend in tech: the locked-in, "grindset" style of work. Founders are embracing the '996' schedule — 9 a.m. to 9 p.m., six days a week — and cutting out alcohol and sex. But Laqua takes it to an extreme. Here's what he told Stebbings on the podcast: He isn't sleeping much Laqua said he has a mattress on the floor of the Corgi office. His employees call it "Nico's room." "I don't spend every single night there anymore," he said. "I used to shower at the Equinox one street over, and they close very early, like 8 p.m. on Fridays. So, that was unpleasant." Other startup founders are also embracing the office bed. When Business Insider toured young founders' apartments in September, several pointed to couches and blow-up mattresses in their offices. While the fusion of work and life may work for some, it can also cause burnout. Laqua also said that he isn't sleeping much. He said that he gets three to four hours of sleep a night. "I would rather measure my lifespan in victories than years," he said. The 7-day workweek Some leaders want a four-day workweek. Laqua aims for seven. "Whatever you can get done in five days, I promise you you'll get more done in six and seven," Laqua said. "You should go all out." Laqua said that high-growth startups in San Francisco have full offices on the weekends. "I don't think it's a coincidence." That doesn't mean Corgi workers can't take a rest day. Laqua said that its employees take a day off "every now and then" — but that they don't have a ritualized weekend of rest. "If your days off happen to be Saturday and Sunday every week, then you will not have a place at Corgi," he said. Linear cofounder Karri Saarinen wrote on X that Laqua's thinking represented that of young founders "where the startup becomes their identity." "They have a hard time doing anything else, and cannot understand that your work is not the person that is you," Saarinen wrote. "But activities outside of work can grow you as a person too and make you do better work." Laqua responded: "If you're obsessed with a problem, you work hard." Those doggy tattoos When Stebbings shared the interview online, he included one extra detail: that "2/3 of the first 30 team members have the Corgi logo as a tattoo." The statistic is not discussed in the interview. In September, Laqua told the Wall Street Journal that "two-thirds of our early employees got Corgi tattoos." Corgi has 177 employees, per PitchBook. Laqua and Corgi did not respond to requests for comments from Business Insider. The tattoos stirred up debate online. "Imagine getting a tattoo all for building B2B SaaS," wrote Coinbase's Richard Wu. Former Lovable engineer Tiger Abrodi called it a "clown show." Others seemed intrigued. OpenClaw chief architect Vincent Koc quoted the line on X, added "Hmmmm," and tagged the agent's creator, Peter Steinberger. On the podcast, Laqua emphasized the importance of strong branding. "The cosmetic stuff does matter," he said. "There's a reason why governments and religions and all of the really important things tend to care about symbols." Would you work at a company with the expectation of a seven-day workweek? Let us know below: Read the original article on Business Insider

Rohan Prabhu / Step Security Blog: Researchers find packages in the @redhat-cloud-services npm namespace shipped malware that harvests credentials for GitHub Actions, AWS, GCP, Azure, and others  —  Several packages in the @redhat-cloud-services npm scope were found to carry malicious payloads that fire via a preinstall hook on every npm install.

Discussion | Link

The author has been living with alpha-gal syndrome for 18 months. The condition, usually caused by a tick bite, prevents him from being able to eat pork and red meat. Courtesy of Zack Barnes. I was having frequent bouts of what I thought was food poisoning. A blood test revealed that I had alpha-gal syndrome, which causes an allergic reaction to meat. The condition, often caused by a tick bite, has caused me to change my routines drastically. When I broke out in hives on a Friday afternoon, I didn't know what was going on inside my body. This wasn't the first time something like this happened, so I knew all the questions the doctor would ask: New detergent? No. New lotion? No. New soap? No. New food? No. When it came to new foods, I had consciously decided to stay away them in recent months. This restriction came about because I kept being jolted awake in the middle of the night with what I thought was food poisoning. These frequent sicknesses made me start to worry about food safety. When I did eat out, I always made sure the meat was fully cooked before taking a bite (which is especially challenging because I am colorblind). At home, I religiously used a meat thermometer when cooking to make sure the meat was cooked through. I always wash my hands after touching meat, and I never let food sit out for longer than two hours. No matter the steps I took, I was still getting sick. With the arrival of hives, I was determined to figure out what was going on. After sending in pictures of my stomach full of hives to my dermatologist's office, I was ordered to take a blood test. After that, I finally got my diagnosis: alpha-gal syndrome. What is alpha-gal syndrome? Alpha-gal syndrome is a meat allergy often caused by being bitten by a lone star tick. I live in Tennessee, where getting tick bites has been common my whole life. This tick bite likely happened when I was cutting my grass. Unlike other food allergies, people with alpha-gal often do not have a reaction until several hours later, when food is being digested. That delayed reaction is why it is so hard to diagnose alpha-gal without a blood test. Looking back on that afternoon when I developed hives, I remember having had a burger for lunch. By the time I got home for work that day, the allergy had set in, and hives were appearing on my skin. The author (not pictured) suspects he was bitten by a tick while cutting his grass, which led to him contracting alpha-gal syndrome. Andrija Nikolic/Getty Images This syndrome is becoming more common I was so happy to have a name for what was happening to me and I've learned that am not alone in experiencing these symptoms. According to the CDC, as many as 450,000 Americans might have alpha-gal, with many people not realizing they have it. The number of cases is expected to continue to rise as more people become aware of alpha-gal and lone star ticks spread across more parts of the United States. At the end of May, the Department of Health and Human Services announced its intention to focus on reducing tick-borne illnesses, such as alpha-gal and Lyme disease. With cases on the rise, the US Department of Health and Human Services announced its intention to focus on reducing tick-borne illnesses, such as alpha-gal and Lyme disease, in late May. gabort71/Getty Images What it's like to live with alpha-gal Compared to others with alph-gal syndrome, I have it pretty easy. Some people are extremely sensitive to any type of mammalian byproduct, such as gelatin or milk. For me, I can still drink milk, eat ice cream, and be exposed to those byproducts. I'm also lucky that when I have an allergic reaction, it just feels like I have food poisoning. While it is extremely uncomfortable, I have not experienced any type of anaphylaxis. However, that is always a risk every time I ingest my allergen, so I've been told to carry an EpiPen. I was diagnosed about 18 months ago, and I'm still getting used to keeping it with me. I'm trying to do a better job of remembering to bring it whenever I leave the house. I no longer get hives when symptoms occur; now I just feel like I have food poisoning, which I treat with rest and hydration. After 24 to 48 hours, I am usually back to normal. I rarely eat outside of my home now The hardest part of living with this diagnosis has been my inability to eat outside my home. I have learned the hard way about the potential for cross-contamination at restaurants. There are now only a handful of places I order food from, and most of those are fast-food places. Fast food places usually cook the food the same way across their locations, so it's easier to find out if the air fryer they use is also being used for any type of red meat. Fast food places also do a great job with their allergen menus that you can easily find online. In some ways, being diagnosed with alpha-gal has made my life better. I was forced to create new routines at home to drastically increase the number of homemade meals and reduce eating out. Creating new routines is something I especially struggle with. It takes a great deal of executive function and self-regulation to both plan ahead and follow through. I had told my wife for years that I would overhaul how we plan and make dinner, but I never did. As someone who studies executive function for a living, I knew that getting a new routine started would make my life easier in the end. Being forced to do it was the push that I desperately needed. It gave me the motivation to finally create a system. Now, we eat almost exclusively at home. This allows me to control all the ingredients we're consuming. I've become great at reading food labels. I already had practice because our kids have dealt with food allergies, but I've really had to pay attention to ingredients because red meat is not required to be listed as an allergen on packaging. Sometimes lard is in refried beans, tortillas, and pie crusts. So I am always on the lookout for ingredients that might trigger a reaction. I've stepped up my cooking and baking game. For cooking, we strive to keep the freezer stocked with precooked chicken and ground turkey that can be easily added to a bowl with rice and vegetables. So when those times come when we would have ordered out, I can quickly defrost the meat and have a meal ready for my wife and me. This has been the most helpful as dinner time is always chaotic with two little kids running around. Now that I can no longer eat ground beef or pork, we get all of our protein from chicken or turkey. When we're looking for a burger, we eat turkey or chicken burgers. We quickly learned that imitation burgers do not taste good to us. My life is different, but not bad What started out as a huge inconvenience, and it still is, has led to some changes in my life that I'm proud of. Creating new routines is hard, especially while raising young children. It takes a while from starting a new routine to making the routine stick. But these new routines have made a huge difference in my life. Thanks to the steps I have taken, I have not had an alpha-gal episode in 6 months. I've continued to learn about this disease, and I'm told that there's a chance that my allergy will go away over time. If it does ever go my way, I don't think I would change the good habits I have created. Read the original article on Business Insider

From CFD and FEA to digital twins, carmaking now involves a lot of virtualization.

Jason Koebler / 404 Media: Hackers say they used Meta's AI support chatbot to change emails tied to Instagram accounts, amid a wave of high-profile account takeovers; Meta fixed the issue  —  The exploit shows the extreme risk of offloading technical support to AI.  —  Hackers say that they used Meta's AI support chatbot …

Get the most out of each coding model to have a very powerful coding setup The post How to Combine Claude Code and Codex for Maximum Coding Power appeared first on Towards Data Science.