Latest news with #Delphi
%3Amax_bytes(150000)%3Astrip_icc()%2FTAL-header-athens-greece-hannah-howard-EDTRAVESSAY0725-eaf0f0594a1841a29a52dc5d2fa7d009.jpg&w=3840&q=100)
Travel + Leisure
2 days ago
- Health
- Travel + Leisure
An Eating Disorder Got in the Way of My Travel Dreams—but My Recovery Helped Me Reclaim the World
There was a time when I couldn't imagine getting on a plane, let alone biting into a croissant in Paris or swimming in the Mediterranean. Or rather, I could imagine those things, but they lived in a fantasy life, far removed from the one I actually inhabited. That life belonged to a thinner, more disciplined version of me—someone who had finally earned joy, earned adventure, earned a seat at the table. I punished my imperfect body with a revolving door of diets and restrictions that crescendoed into a full-blown eating disorder: first, anorexia; then binge eating; then a confusing diagnosis of EDNOS—eating disorder not otherwise specified, a name that felt as vague and dissatisfying as the experience itself. Hannah while in Delphi, Greece, famous for its ancient oracle of Apollo. In college, I dreamed of studying abroad. My friends posted pictures from Florence and Buenos Aires, sipping wine and hiking ruins, glowing with possibility. I told myself I couldn't go because of credits or timing. But deep down, I knew the truth: I was afraid. Afraid of unfamiliar food. Afraid of being the biggest girl on the trip. Afraid of what might happen if I wasn't in complete control. My eating disorder had one job: to keep me small. And it did. Not just physically, but in every way. It whispered, After you lose the weight, real life can begin. And I listened. In my early 20s, I entered treatment. I met other women—some my age, others in their 30s, 40s, 60s—who had lived their whole lives in the shadow of that same lie. They hadn't taken the trip, or eaten the cake, or worn the dress. They were still waiting to be 'ready.' And I saw, with a kind of clarity that felt like heartbreak, how long a person could wait. Recovery, for me, was not linear or pretty. It was messy, slow, and filled with backslides and tiny wins. But as I stitched myself back together, piece by painful piece, I made myself a promise: I would not live the rest of my life in a cage of my own making. A few years after treatment, I booked a trip with a friend who got it to Mexico. I cried in the airport bathroom. I second-guessed every outfit in my bag. I worried I was too big to deserve adventure. But I got on the plane anyway. And then something wild happened: I had a good time. A really good time. I ate street tacos without counting calories. I stood in front of ancient ruins and felt small in the best way possible. I wore a bathing suit on a beach where no one knew me, and I took photos where my body wasn't hidden or strategically angled—and tried (sometimes successfully, sometimes not) not to let the critical voice in my head narrate them. That trip cracked something open in me. It was the first time in a long time I felt connected to my body—not as a problem to fix, but as a vessel that carried me somewhere new. Since then, I've traveled to dozens of places I once would have ruled out entirely. I've hiked through olive groves in the south of Spain and floated in thermal waters in Costa Rica. I've tasted jackfruit curry in a banana leaf hut on the backwaters of Kerala and dumplings in a Bangkok night market. I've also had meltdowns in hotel rooms, cried after seeing an unflattering photo, and ordered room service because I was too overwhelmed to deal with the buffet. Recovery is still a process. The voice in my head didn't disappear—it just got quieter. And I've gotten better at not believing it. Traveling on a boat while in Istanbul. Travel is, in many ways, the opposite of my eating disorder. My eating disorder was about control, rigidity, sameness. Travel is about surrender, openness, surprise. It asks me to be present in a body I don't always love, but am learning to live in with more compassion. In Istanbul, I stepped into a Turkish hammam, naked but for a towel and nerves. I expected judgment, comparison. What I found instead were women of every shape and size, laughing, scrubbing, stretching. Nobody cared about my thighs. In Switzerland, after what felt like 10 straight meals of potatoes, bacon, and cheese, the old panic started to bubble up—you've ruined everything. But I reminded myself: I am OK. I am resilient. I will eat a vegetable again. Every time I travel, I reclaim a little more space—not just in the world, but in myself. I take up room in photos. I order the special even if it's rich. I walk through cities without shrinking. I don't wait to be 'ready.' I go now. This isn't a story about being healed. It's a story about choosing presence over perfection, again and again. Sometimes when I travel, I still catch myself worrying about how I look. But then I try to shift the question: How do I feel? Am I alive? Curious? Awake to the smells of a new market, the salt on my skin after a swim, the warmth of bread pulled fresh from a bakery oven? That's what I want to remember. That's what I want to collect. Not proof that I looked thin enough in a photo, but memories that I was there—fully, messily, joyfully there. My body is not a before-and-after story. It's a passport. And I intend to keep using it.
Yahoo
3 days ago
- Health
- Yahoo
Study reveals best way to communicate link between alcohol and breast cancer
A new study has revealed the best way to communicate the link between alcohol and breast cancer. The research, undertaken by Oxford Brookes University which has a campus in Swindon, and funded by the charity Prevent Breast Cancer, focused on women aged 40 to 65 in the UK. It found that many women in this group were unaware of the connection between alcohol consumption and breast cancer. Read more: 250-year-old Wiltshire estate launches summer adventure series The study, titled 'Rethinking the message on alcohol and breast cancer with UK women: a Delphi study', was published in the journal Health Promotion International. It involved a three-stage process, which began with a survey of 260 women, followed by seven online focus groups and a collaborative workshop. The study's lead author, Dr Emma Davies, said: "We often think of alcohol as causing liver disease, but there's plenty of research showing that drinking alcohol can lead to seven types of cancer, including breast cancer. "Evidence shows that people who are aware of the link between alcohol and cancer are more supportive of stronger and more effective alcohol policy. "This means that raising awareness isn't just about individual behaviour change, it is about changing how we think about alcohol at all levels of society." The study found that several factors, including cultural norms, mistrust of official messaging, psychological defence mechanisms, and stigma, reduced the effectiveness of health warnings. Fear-based messaging was also found to be counterproductive, as it often led to denial rather than proactive change. Dr Davies said: "It's clear that fear, blame and shame don't work when it comes to raising awareness of the risks associated with drinking alcohol. "Cutting back on alcohol can help to reduce the chance of getting cancer, but can also give us plenty of other benefits, such as better sleep and improved mood." The study concluded that narrative-based framing, using personal stories from peers who have experienced breast cancer, was more effective than stark statistics or scare tactics. Messages were most accepted when framed positively, highlighting how reducing drinking can empower women and protect their health, rather than through guilt or blame. Dr Davies added: "Importantly, we need a clear and evidence-based alcohol policy to reduce risks across the population. Read more: WHSmith shops to be renamed as firm sells up at cut price "We need to understand why people drink and what the emotional and cultural barriers are to giving up or cutting down. "We hope our study will equip policymakers, charities, clinicians, and health communicators with an evidence-based roadmap to reshape prevention campaigns and reduce alcohol-related harms, including breast cancer and other cancer cases." For more information and advice on alcohol and cancer, visit the World Cancer Research Fund's Cancer Prevention Action Week page.
CNBC
5 days ago
- Business
- CNBC
Cliff Asness' AQR sees multiple hedge funds up double digits in 2025, beating the market
AQR Capital Management took advantage of a volatile first half of 2025, with a duo of hedge funds doubling the S&P 500's return. The Apex strategy from Cliff Asness' firm, which combines stocks, macro and arbitrage trades and has $4.3 billion in assets under management, rallied 11.4% in the first six months of the year, according to a person familiar with AQR's returns who asked to be anonymous as the information is private. AQR's long-short Delphi equity fund, with $4.1 billion in assets under management, gained 11.6% net of fees in the first half of 2025, the person said. The stock market staged a stunning rebound this year even as uncertainty remains amid an aggressive trade war and Middle East escalation. The S&P 500 has rebounded from a near 20% sell-off in April, going on to score a new record high on Friday and again on Monday. The equity benchmark is up 5.3% year to date. AQR's alternative trend-following Helix strategy has returned 7.4% so far this year, the person said. Asness co-founded AQR in 1998 after a stint at Goldman Sachs. He and his partners established the quant-driven firm's investment philosophy at the University of Chicago's Ph.D. program, focusing on value and momentum strategies. The firm has successfully expanded into multistrategy approaches in recent years. AQR has $142 billion in assets under management, up from about $99 billion at the start of 2024. AQR declined to comment.
Yahoo
25-06-2025
- Science
- Yahoo
New space startup Lux Aeterna wants to make satellites reusable
Satellites can accomplish incredible tasks like provide internet, or help monitor wildfires. But many of them ultimately meet a fiery death burning up in the Earth's atmosphere. Others use their last bit of fuel to move to so-called 'graveyard' orbits, where they circle the planet in a perpetual deep freeze. A new startup called Lux Aeterna wants to change this. The Denver-based company, which is coming out of stealth today, has designed a reusable satellite called Delphi that it aims to launch — and land — in 2027. If successful, it could help slash the cost to get satellite payloads into space. It would also make the process far more flexible than it is today, since satellites are designed to stay in orbit for years and essentially can't be modified for other uses. These attributes have already piqued the interest of the Department of Defense, which has made low-Earth orbit an important part of the United States' military strategy. Venture capitalists have also taken notice — and written checks. The startup's pitch was attractive enough to generate $4 million in pre-seed funding, led by Space Capital and including other early-stage funds like Dynamo Ventures and Mission One Capital. Founder and CEO Brian Taylor said the idea for Lux Aeterna came to him last year while he watched his former employer, SpaceX, launch one of its Starship test vehicles into space. 'I want to fill Starship with something amazing, and something that changes the entire industry,' Taylor recalled thinking during an interview with TechCrunch. Starship is the biggest rocket ever built. As part of that, it has the potential to send larger payloads into space than was previously possible. Size matters for people who build satellites and other spacecraft, since they're often working backwards from the simple constraint of what can fit inside a rocket's cargo area. And Starship is not alone — there are other heavy-lift rockets in the works, too, like Blue Origin's New Glenn. It's hard to design a satellite that can survive the brutal forces of re-entering the Earth's atmosphere at high speeds. But with the extra space afforded by heavy-lift rockets, Taylor said it's possible to build one that can survive multiple re-entries without having to compromise on the technology because of cost or weight tradeoffs. In the case of Lux Aeterna, that means using a heat shield. In the rendering the startup released Wednesday, the Delphi satellite's conical heat shield is reminiscent of the ones that protected some of NASA's most famous spacecraft. There's a reason for that, according to Taylor: Those designs worked. 'We definitely looked at what NASA had done in the past on exploratory missions [and] sample return missions, and that really helped justify the architecture that we've gone with,' he said. 'I think it's very important, when you're doing something ambitious like this, that you're not reinventing the wheel on everything, right?' Taylor declined to get into further specifics about how the Delphi satellite will work, or how Lux Aeterna will refurbish the craft between launches. (The design appears to involve the ability to fold the satellite bus structure so that it fits safely behind the heat shield.) To be sure, he has plenty of experience in the satellite world. In addition to working on Starlink at SpaceX, Taylor also worked on Amazon's Kuiper satellite program, and at space infrastructure startup Loft Orbital. The plan for Delphi is to launch on a SpaceX Falcon 9 rocket in 2027, perform a full orbital flight, and then come back down to Earth. Then Lux Aeterna wants to do that all over again to prove out Delphi's reusability. From there, Taylor said his team is working on a larger production vehicle that will demonstrate far greater reusability. Despite decades of spaceflight innovation, Taylor said he believes the industry is still very young, which leaves plenty of opportunity for a company like Lux Aeterna to establish a long-running business. 'It's not to the maturity level of [computer] chips. It's not at the maturity level of automotive,' he said. Satellite reusability will help change that. And while Taylor is committed to that cause, he said he's thrilled about all the things he can't imagine that will exist in a space-based economy. 'We don't know what we don't know is going to come,' he said. 'That's probably the most exciting part.' Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
Fast Company
24-06-2025
- Business
- Fast Company
Meet Delphi, the AI startup that lets experts turn themselves into chatbots
Everyone who's ever talked to ChatGPT, Claude, Gemini, and other big-name chatbots recognizes how anodyne they can be. Because these conversational AIs' creators stuff them with as much human-generated training content as possible, they don't end up sounding like anyone in particular. Instead—to borrow the title of a 1986 book by philosopher Thomas Nagel —they offer the view from nowhere. But what if you could train a bot by feeding it material you'd created, reflecting your knowledge, way of thinking, and style of self-expression? Instead of sounding like nobody, it might sound like you, or at least a rough approximation thereof. Given enough training fodder and sufficiently advanced technology, such a bot could even be capable of serving as an automated extension of your own brain. That's the idea motivating Delphi, an AI startup whose 14-person team is building a platform for what it calls 'digital minds.' The 2½-year-old company, which previously raised $2.7 million in seed funding, is announcing its $16 million Series A round, led by Sequoia Capital with participation from Menlo Ventures, Anthropic's Anthology Fund, Michael Ovitz's Crossbeam, and others. It will use the new infusion of cash to continue to build out its web-based tool kit, which already includes features for creating, refining, and monetizing digital minds. The creators currently using Delphi tend to be people with substantial existing followings they've built up through websites, newsletters, podcasts, social-media accounts, speaking engagements, books, and other modes of communication. They include business-advice newsletter kingpin Lenny Rachitsky, wellness coach Koya Webb, HubSpot CEO Brian Halligan, sex therapist Vanessa Marin, motivational speaker Brian Tracy, financial adviser Codie Sanchez, bodybuilder-actor-governator Arnold Schwarzenegger, and many others. It's possible to chat with their digital minds via a texting-style typed session or a voice call; creators can also enabled video calls. Despite the voice and video options, the core of the concept isn't about deepfaking how a creator looks and sounds. 'We've centered everything around the mind,' Delphi cofounder and CEO Dara Ladjevardian told me during my recent visit to the company's San Francisco office. That 'doesn't mean just capturing your expertise, but also capturing how you reason about things,' he says. 'So you can give personalized advice, so we can be predictive of what you might say in new situations.' Even if Delphi's emphasis on the quality of the conversation over audiovisual razzmatazz helps it avoid a disconcerting uncanny valley effect—'We've seen consumers don't really care about the video,' Ladjevardian says—it's still a bit of a mind-bending proposition. Along with overcoming the obvious technical challenges of teaching AI to channel a specific person in a way that's trustworthy enough to actually be useful, Delphi will also need to get consumers comfortable with seeking advice from simulated versions of real experts. 'This, today, I think, to a lot of consumers just seems weird,' says Sequoia partner Jess Lee, who led the firm's investment in Delphi. 'We need to cross the chasm and there need to be more people using them. And I think that will come with new Delphi owners shipping and showcasing what it can do.' Already, Delphi is helping early adopters scale up their ability to engage with audiences. 'We've always had a fundamental challenge, which is that more people want to ask me questions than I'm possibly able to get to in 10 lifetimes, let alone in the next year,' says relationship and confidence coach Matthew Hussey. Last year, his company created Matthew AI, a Delphi digital mind trained on 17 years of his existing content. By calling it 'Matthew AI,' he hoped to manage expectations about what it could and couldn't do. Even then, he wasn't sure how customers would respond. 'We sort of launched this squinting, bracing ourselves for a whole bunch of mixed reviews,' Hussey told me. And it's probably been one of the most well-reviewed things we've ever created.' How to create a (digital) mind In Ladjevardian's account, the Delphi story begins with a gift he received in 2014: a copy of Ray Kurzweil's book How to Create a Mind. In it, the noted inventor and futurist explored the inner workings of the human brain and how they might be re-created in computer software. As it often does, Kurzweil's own mind had raced ahead of what was possible at the time. But the forward-looking analysis resonated with Ladjevardian. He eventually started an AI company that let people shop by sending text messages, then quickly sold it. Entrepreneurship ran in Ladjevardian's family: Decades earlier, his grandfather had been a successful businessman in Iran. After the 1979 revolution, 'he had to be smuggled out of the country, came to the U.S. with nothing, and was able to build a life,' explains Ladjevardian, who'd started his AI shopping company on his own, found life as a solo founder lonely, and craved wisdom from his grandfather. But a stroke had greatly limited the elder man's ability to communicate. Ladjevardian's thoughts turned back to Kurzweil's book. 'He talks about the mind being a hierarchy of pattern recognizers,' Ladjevardian says. 'And when I was building this first startup, I realized an LLM is pretty much a pattern recognizer. So I set out to create a digital mind for my grandpa and use it for advice. It was therapeutic.' In November 2022, the experience of turning a memoir his grandfather had written into an interactive tool led him to start Delphi with Sam Spelsberg, a colleague from Miami-based OpenStore, where Ladjevardian had worked after selling his startup. Spelsberg is now Delphi's CTO. Harnessing AI to preserve human insight for the ages remains part of the story at Delphi, whose website calls the service 'your path to digital immortality.' But by applying the technology to the immediate needs of people who make a living as experts on various topics, the company gave itself a mission with a clearer business model. It offers free accounts trained on 100,000 words and limited to text chatting. Creators who pay $79, $399, or $2,499 per month get progressively richer access to features such as larger training sets, voice and video calling, analytics, setup help, and the ability to charge for sessions and keep 85% or more of the proceeds. (Delphi is already realizing revenue from its cut but doesn't disclose a specific figure.) Creators decide how much free access users get to their Delphi experience and when a paywall kicks in. As Sequoia's Lee points out, there are additional ways digital minds can bolster a business, such as upselling products and providing customer support: 'I talked to a clinician who runs a nutrition program and uses it to train other nutritionists on his program,' she says. In my unscientific experiments chatting with a few digital minds, I learned not to expect too much from the technology in its current state. Maybe it will someday pass a sort of specialized Turing test where you're unsure if you're talking to Lenny Rachitsky or his synthetic doppelgänger. For now, however, Delphi Lenny's auto-generated observations are rife with telltale evidence of their artificiality, such as a tendency to repeat the same phrases. Still, the tips it gave me on how cash-strapped startups can hire the best talent seemed solid and included links back to Real Lenny's Substack and podcasts. According to Ladjevardian, Rachitsky uses his Delphi to help shape his writing: 'People can ask him follow-up questions when they read a blog, and he can look at analytics to see what's resonating and use that for ideas for future content strategy.' Even if today's digital minds do churn out responses that feel, well, digital, the originating human's viewpoint can come through. When I asked the digital version of investor Keith Rabois about the ideal place to start a company, it was as blunt and opinionated as the real thing: 'Miami offers a pro-business environment, a growing talent pool, and a lifestyle that attracts top-tier people. . . . San Francisco, on the other hand, is a disaster. It's unsafe, overregulated, and culturally toxic.' ChatGPT would never put it quite that way. (For the record, Delphi itself relocated from Miami to its current digs in San Francisco's Jackson Square neighborhood: 'If you want to attract the best engineers, you've got to be in San Francisco,' Ladjevardian says.) Then there's another digital mind whose answers I found of particular interest. Before I met with Ladjevardian, he'd trained one based on my large archive of published writing for demo purposes. I later supplemented it with additional content until it drew on more than 5,000 items—articles, podcasts, tweets, and more. Placing a voice call to a simulated version of yourself speaking in a synthesized version of your own voice is a surreal exercise, but my biggest takeaway was that tech journalism is not the best source material for Delphi in its current form. In most cases, articles I wrote years ago about now-obsolete products are of limited training value today. And Delphi didn't yet know my take on matters of the moment such as Apple's upcoming VisionOS 26. My conversations with my digital self left me with a greater appreciation for why the experts featured on Delphi's site tend to focus on topics with longer shelf lives, from leadership to sex. Please don't call them clones In a world full of tech companies whose self-professed aspiration to create AI that's smarter than any human, there's something reassuringly down-to-earth about Delphi's short-term goal of helping specific humans boil down what they know into a monetizable product. Yet the startup's work to imbue AI with human-like traits is inherently fraught. When other companies are in the news for attempting to humanize AI, it's often in a negative light, for reasons ranging from the silly (the failure of Meta's terrible celebrity chatbots) to the tragic (lawsuits resulting from teens developing an unhealthy relationship with Details as mundane as terminology matter. Originally, Delphi referred to its AI conversationalists as 'clones,' but that 'sounds a little dystopian when you hear it at first,' says Lee. It 'seems like a facsimile of a person. That's not really what you're doing. You're just taking someone's existing expertise, their blog posts, their tweets, and you're making it conversational.' A Delphi-Sequoia brainstorming session led to the digital mind term, which Lee finds 'somehow much more accessible.' That said, when I spoke with Ladjevardian, he was still getting used to the switch and referred to 'clones' a few times before correcting himself. Even with Delphi's emphasis on practical advice and downplaying of fancy visuals, a lot could go wrong. Ladjevardian says the service doesn't let anyone generate digital versions of other people and is manually vetting users by making them upload photos of themselves holding an ID. (It has, however, resuscitated some long-deceased philosophers and other notable figures; I chatted with one former president of the U.S. whose greeting—'Hi, I'm Abraham Lincoln. How can I help?'—was a touch out of character, though he sounded more Lincolnesque in the conversation that followed.) The company also has guardrails in place to prevent inappropriate answers: When I asked physician Mark Hyman's digital mind questions involving my own health, it did not attempt to answer them and instead recommended that I see a healthcare provider. Ladjevardian, who volunteers that his project to build a bot based on his grandfather got him 'canceled on Twitter,' understands the need to acclimate people to what Delphi is doing. Some companies that have unsuccessfully pursued vaguely similar ideas 'were founded by AI researchers who were way too focused on the technology,' he says. 'And this is a very human company. I've had people cry to me after creating their digital mind.' As the startup sees where its product can go, creating experiences that can bring people to tears will be optional. Engendering confidence—even among those prone to skepticism about AI—won't be. Ladjevardian's bet: The fact that garden-variety LLMs have left us awash in information and advice of questionable provenance makes Delphi's association with specific human experts only more powerful. 'Whenever there's an era of abundance, the pendulum swings and people want curation and trust,' he argues. Even the brightest of digital minds might have trouble foreseeing whether that theory will indeed play out to the company's advantage. The final deadline for Fast Company's Next Big Things in Tech Awards is Friday, June 20, at 11:59 p.m. PT. Apply today.
