SCROLL NEWS / DISCOVERY
Search headlines
Find the stories shaping the conversation.
Results for " mist " 2,093 found 🔀 AI-powered Shuffle
Bills urged to roll the dice in NFL Draft
The Bills might be looking to fix their Keon Coleman mistake.
LSU to Open $148M Science Building This Fall
A 148,000-square-foot facility set to open in the 2026-27 academic year will have research and teaching laboratories with technology supporting the biology, chemistry, geology, math, physics and astronomy disciplines.
ID@Xbox Presented by IGN April 2026 Showcase Announced: Here's How to Watch and What to Expect
IGN is proud to partner once again with ID@Xbox for another exciting showcase on April 23, 2026, that will feature such indie games headed to Xbox and PC as Mistfall Hunter, There Are No Ghosts at the Grand, Aphelion, and Solo Leveling Arise Overdrive. Here's everything you need to know and how you can watch this showcase with us.
The US in Brief: Fuelling economic worries
Our daily political update, featuring the stories that matter
Singer-songwriter John Gorka talks early inspiration, latest album
John Gorka’s newest album, “Unentitled,” marks his return after nearly a decade, showcasing fresh, optimistic tracks that reflect his belief in music’s healing power despite industry challenges.
OPINION: Tribal and rural healthcare at risk in Gov. Stitt’s attack on Medicaid
Oklahoma leaders often visit Cherokee Nation’s Health System and celebrate our progress and tout our unmistakable impact on the health and economy of rural communities.
Computer science was once a 'golden ticket' — now it's seeing the biggest enrollment drop of any major in 6 years
Nobel Prize–winning economist Simon Johnson said AI has "substantially wiped out" coding as a reliable source of opportunity.
Computer science was once a 'golden ticket' — now it's seeing the biggest enrollment drop of any major in 6 years. What students are choosing instead
Nobel Prize–winning economist Simon Johnson said AI has "substantially wiped out" coding as a reliable source of opportunity.
Family of 7 took a multigenerational trip to Paris
My family of 7 took a multigenerational trip to Paris. We made a few mistakes and realized that paying for convenience was almost always worth it.
Curing U.S. Health Care, Part I
In 2008, much to their own surprise, leading Democrats unified around a program of major health care reform. Policy wonks had spent years developing the concepts behind what eventually became Obamacare; big Democratic victories in the 2006 midterms and the prospect of controlling both Congress and the presidency made it possible to imagine turning those ideas into reality. During the Democratic primary, Hillary Clinton and Barack Obama advocated similar plans, based on those ideas, for expanding insurance coverage. And it happened! The Affordable Care Act was enacted in 2010. When it was fully implemented in 2014, millions of Americans got health insurance: Impressive as the raw numbers are, they don’t tell the whole story. Before the ACA, even upper-middle-class Americans often found it impossible to get health insurance if they had pre-existing medical conditions. Many Americans were trapped in jobs they wanted to leave but couldn’t for fear of losing their employment-based coverage. Meanwhile, dire predictions from the usual suspects about runaway costs proved wrong. In fact, overall U.S. medical spending has grown much more slowly since the ACA was enacted than before. But the U.S., alone among advanced nations, still falls far short of providing universal health care. As you can see from the chart above, 8 percent of the population was still uninsured in 2024, a number that is set to rise over the next two years as a result of Republican policies. True, many of the uninsured in 2024 were undocumented immigrants, who we don’t try to cover. But there are still a lot of uninsured. Moreover, a significant number of Americans who have health insurance are in fact underinsured. As a result, they are at risk of incurring devastating healthcare costs and are sometimes forced to forgo needed care. This number is set to rise sharply in the next two years as a result of Republican policies adopted under Donald Trump. And not only is the U.S. unique among advanced countries in its under-provision of health care coverage, it also incurs by far the world’s highest healthcare costs per capita. So now may well be a good time to get behind a new push for major health reform — an effort, if you like, to finish the job begun under Obama. Today’s primer is devoted to the economics of health reform. During his failed effort to repeal the Affordable Care Act, Donald Trump famously complained, “Nobody knew healthcare could be so complicated.” Actually, we did know — and it’s not that complicated. Health economists understand the principles very well. And because health policy varies greatly among advanced nations, we know a lot about what works and what doesn’t. This will be the first in a series about health reform. Beyond the paywall I’ll address the following: 1. Why markets can’t be trusted to deliver healthcare 2. Routes to universal healthcare 3. What works? In a follow-up post I’ll discuss the pros and cons of different approaches, and possible paths forward for the United States.
‘The current system right now is unsustainable’: top economist sees a crucial crack in the economy
“The current system right now is unsustainable,” one economist said. As baby boomers age, the problems may only intensify.
Pope Leo’s target is bigger than Trump. He’s rebuking the ‘MAGA Jesus’
If you think the clash between Pope Leo XIV and President Donald Trump is just a war of words that will pass when the news cycle moves on, you are most likely mistaken.
Eckerd students present research at scientific conferences around the world
This spring, Eckerd students travelled around the world to present their research at conferences. In February, marine science students attended the Ocean Sciences Meeting in Glasgow, Scotland. In March, chemistry
Rosalind Franklin, chemist: “Science and everyday life cannot and should not be separated”
A pioneering DNA scientist reflects on science, faith and purpose, offering a grounded view of how discovery and everyday life intertwine
Scientists Raise Concerns Over Newly Recognized Pollutant Found Everywhere in the Air
Scientists have uncovered unexpectedly high levels of methylsiloxanes in the air worldwide, suggesting a hidden and widespread form of pollution.
This May Be Why Eric Swalwell Thought He Could Get Away With It
Party leaders treated him like a man who could do no wrong. That was ... a mistake.
A step toward stability? Atlanta health professionals react to CDC nomination
Dr. Erica Schwartz must still be confirmed by the Senate. Metro Atlanta health professionals and former CDC employees are ‘cautiously optimistic.’
Diamondbacks capitalize on Blue Jay mistakes, win series opener
D-Backs capitalized on Blue Jay mistakes and walked out victors in their return home.
Council demands more transparency from Mayor Mamdani on mental health response program
The City Council on Thursday passed legislation that requires the mayor’s office to submit regular reports tracking the performance of B-HEARD.
Why Political Pressure on Central Banks Can Raise Your Borrowing Costs
The question of central bank independence came up during a fireside chat on Grounds with Philip Lane, chief economist of the European Central Bank, hosted by economist and Darden professor Frank Warnock.
Secretary Kennedy Appoints Casey B. Mulligan as HHS Chief Economist and Regulatory Officer
HHS appoints Casey B. Mulligan, PhD, as Chief Economist and Chief Regulatory Officer.
5 mistakes to avoid on your first long-distance Amtrak ride
I took the 20-hour Amtrak route between Chicago and New York. These five mistakes kept me from enjoying it more.
The US in Brief: ICE chief steps down
Our daily political update, featuring the stories that matter
Glenwood Park businesses, residents optimistic over new Beltline expansion
The latest Beltline expansion is sparking hope among Glenwood Park residents and business owners, who anticipate what new opportunities it could bring.
‘Save the Garden’ turns to embattled lawyer as petition effort falters
Sabatini claims a mistake that led to the petitions' rejection shouldn't be cause for dismissal, but another legal expert disagrees.
Solo cruise surprises on Royal Caribbean's Wonder of the Seas
I took a solo cruise on board Royal Caribbean's Wonder of the Seas and was surprised by how many people mistook me for a lost child.
In Mamdani’s NYC, sewer socialism is out. Pothole politics is in.
The mayor's governing philosophy has gotten a rebrand.
Tumour cells use a genetic trick to become drug-resistant
The same trick could also be used against them
UNC Hussman offers new sports play-by-play class | UNC Hussman School of Journalism and Media
This story, written by UNC Hussman alumna Caroline Daly ’24, was originally published on unc.edu. The UNC Hussman School of Journalism and Media has a new, hands-on class available for students interested in sports careers. Students in “MEJO 390: Sports Commentary and Play-By-Play” can call live radio broadcasts for Carolina sporting events, streamed live on the Go Heels website. This spring, 11 students are calling mostly softball games to gain experience and learn the preparation needed for a successful broadcast. The class is capped at a small number to provide consistent, individual feedback. Matt Krause, the radio voice of Carolina women’s basketball, teaches the class to help students sharpen their skills, using his experience with the Tar Heels and calling games for ACC Network and ACC Network Extra. “There is no substitute for live reps, and that is the only way you get better. The classroom component teaches you how to get better, but going about that requires sitting in that chair, making mistakes and being able to know your broadcast is going somewhere,” Krause said. “These students are doing real live games, and they will have tape where if they want to pursue this. They will have ready-made demos to send for jobs. Being able to utilize all the resources of UNC, GoHeels Productions and the Carolina Athletics brand and create these opportunities for students is really impactful.” In the class, students learn fundamentals like how to announce starting lineups for different sports and how to conduct research before games. Krause has led the class in mock preparation exercises, looking at game charts with information about the different teams and players that they can refer to at pivotal moments during the game. Krause and the students then critique each other, listening to clips from the past week to find ways to improve. It’s a rare class without a slew of quizzes or assignments. Their main tasks are calling games and making a portfolio or demo reel of their best calls at the end of the semester, all with the goal of being ready to step into similar roles after graduation. Laible live streams her coverage on the GoHeels website. Calling a live game while broadcasting her coverage allows her to get hands on experience. (Jon Gardiner/UNC-Chapel Hill) Junior Anna Laible ’27 wished for a class just like this. Her first year at Carolina, she interned with the Coastal Plain League, where she was introduced to play-by-play for baseball. “I enjoy the challenge and uncertainty of each game being different because each sport looks a little different,” said Laible. “Each game is different — not just with teams but what could happen with certain records being broken or random trends. That’s what I enjoy most.” She wanted more opportunities for reps during college to prepare for a career after Carolina, so she pitched an idea for the class to Livis Freeman ’99, distinguished associate professor of sports communication. Laible and Freeman reached out to GoHeels Productions for a partnership and then connected with Krause. “This is how at UNC Hussman we separate ourselves: Our students go into the field ready,” said Freeman. “The biggest compliment we get it is when our students start their careers, they don’t need any hand-holding.” Laible calls a Carolina softball game at Kessing Stadium – after pitching the idea to Hussman faculty. Her interest in sports commentary allowed her to apply her interest as she attends Carolina. (Jon Gardiner/UNC-Chapel Hill) Laible is now a student in the class, which has helped her strengthen her skills and deepen her passion for the field. The goal is to expand to other sports like soccer, lacrosse and field hockey. “The hands-on experience I’ve gotten here is invaluable. I’ve loved the classes and professors because they’ve taught me a lot, but it’s also the things outside of class, too, that have helped me grow,” said Laible. “Since being at UNC, I’ve grown a lot in the versatility, from doing writing to production and commentating. UNC Hussman and Carolina have given me the ability to be strong and well rounded.”
Avoid these common small business tax mistakes
Block Advisors Strategy and Small Business Program Manager, Latasha Randle, chats with Host, Elizabeth Gore, about how self-employed entrepreneurs can simplify taxes, retirement, and healthcare planning. Randle shares practical advice on staying organized year-round, maximizing deductions, and building long-term financial security through proactive planning.
The US in Brief: Trump goes for Powell again
Our daily political update, featuring the stories that matter
With Iran War on Pause, Political Battle Resumes Over Israeli Democracy
A court fight over Itamar Ben-Gvir, an extremist who oversees the police, pits the attorney general against Benjamin Netanyahu and his right-wing government.
The Guardian view on social science research: embracing uncertainty
Editorial: Science rarely produces identical outcomes. Mistaking this for failure turns caution into an excuse for inaction
Carver Academy student takes 3rd in chemistry at state science fair
DOTHAN, Ala. (WDHN) — A student from Carver 9th Grade Academy in Dothan placed high in Alabama’s state science fair. Gabrielle Saffold earned third place in the senior division for chemistry and mater...
Americans stand out internationally for their pessimism about the nation’s political system
People in higher-income countries are generally less likely to say their nation’s political system needs major changes or complete reform. Americans are an exception.
Kennesaw State to launch Ph.D. in Biomedical Sciences in Fall 2027
Building on its leadership in scientific research and education, Kennesaw State University on Wednesday received approval from the Board of Regents of the University System of Georgia to launch a new Ph.D. in Biomedical Sciences. The program will begin in Fall 2027 on the Kennesaw Campus. Housed in the College of Science and Mathematics, the degree will help meet the increasing demand in Georgia and the nation for research scientists in biology, chemistry, biochemistry, and ecology.
Justice Dept. Moves to Vacate Jan. 6 Convictions for Far-Right Extremists
Defending the convictions would likely have required administration officials to assert that far-right groups were acting on behalf of President Trump on Jan. 6, 2021.
Building Custom Atomistic Simulation Workflows for Chemistry and Materials Science with NVIDIA ALCHEMI Toolkit
For decades, computational chemistry has faced a tug-of-war between accuracy and speed. Ab initio methods like density functional theory (DFT) provide high fidelity but are computationally expensive…
Q&A with chemist Max Delferro: Unlocking breakthroughs in catalytic science
In this Q&A, chemist Max Delferro explains how his team is combining catalysis, artificial intelligence and robotics to speed chemical discovery — cutting catalyst development from decades to years while advancing more efficient industrial processes.
University of Arizona launches BS in Science Law, available on campus and online
The BS in Science Law builds a strong foundation in both disciplines through coursework and labs in biology, chemistry and physics, alongside undergraduate law courses.
Q&A with chemist Max Delferro: Unlocking breakthroughs in catalytic science
In this Q&A, chemist Max Delferro explains how his team is combining catalysis, artificial intelligence and robotics to speed chemical discovery — cutting catalyst development from decades to years while advancing more efficient industrial processes.
The US in Brief: Judge bins Trump’s tort
Our daily political update, featuring the stories that matter
Spain Business Confidence Falls 2.3% in 2Q, Most Since 4Q 2022
Spain’s business confidence fell 2.3% in the second quarter compared with the first quarter, the biggest decline since the end of 2022, according to data from the Spanish statistics office. The survey showed 21% of business establishments expect a favorable quarter, while 16% are pessimistic about their business outlook.
Scientists Grow Electronics Inside the Brains of Living Mice
The technology harnesses the brain's own blood chemistry to assemble soft, light-controlled electrodes around neurons.
Exynos 2600 Was Born Out Of Sheer Necessity And Wouldn’t Have Entered Mass Production If Samsung’s Mistakes Not Cost It $3 Billion Last Year
As Samsung races towards self-sufficiency in the chipset race, it only mass produced the Exynos 2600 because it lost $3 billion in 2025
Rep. Eric Swalwell will resign from Congress following sexual misconduct allegations
The California Democrat said in a statement he was “deeply sorry to my family, staff and constituents for mistakes in judgment I’ve made in my past.”
Senate Republican calls proposed NASA budget cuts a ‘mistake’
Republican Sen. Jerry Moran (R-Kan.) on Sunday criticized the Trump administration’s proposed cuts to NASA. “I think it would be a mistake to put money only in the missions related to explora…
Stanford HAI 2026 AI Index: AI posts gains in science and medicine while often struggling to read a clock
Stanford HAI’s 2026 AI Index paints a split-screen picture for R&D. In the report’s science chapter, “frontier models outperform human chemists on average on ChemBench,” yet “score below 20% on replication in astrophysics and 33% on Earth observation questions.” In medicine, the report reaches a similarly mixed conclusion, saying AI is transforming clinical care, but…
"I couldn't protect it" Pete Hines says he left Microsoft-owned Bethesda because he didn't want to watch it be "damaged and broken apart and frankly mistreated, abused"
Former Bethesda head of publishing Pete Hines has spoken about the frustrations which led him to retire back in 2023.
Pirates made a mistake signing Marcell Ozuna over Andrew McCutchen
The Bucs made a bad decision.
The US in Brief: Swalwell’s out
Our daily political update, featuring the stories that matter
At 100 days, Mamdani leans into ‘pothole politics’ and child care push
The mayor focused on child care and worker gains, arguing that basic services are key to rebuilding trust in government.
Scientists Unveil Microscopy Breakthrough That Reveals “Invisible” Molecular States
Scientists have introduced a microscopy technique that reveals a hidden layer of chemistry involving molecules that normally evade detection.
Tribal and Rural Healthcare at Risk in Oklahoma Gov. Stitt’s Attack on Medicaid
Guest Opinion Oklahoma leaders often visit Cherokee Nation’s Health System and celebrate our progress and tout our unmistakable impact on the health and economy of rural communities. Now, Governor Stitt is putting us on a path that will damage that very system. Recently the Oklahoma legislature passed Oklahoma HB 4440 and HJR 1067 to reverse […]
U.S. Moves to Deport Son of Prominent Figure in Iranian Revolution
Federal agents arrested a man whose mother served as a spokeswoman for the Islamist embassy captors during the hostage crisis that began in 1979.
“AI polls” are fake polls
A few weeks after Donald Trump’s second presidential win, I took the train up from London (where I was living at the time) to Oxford to attend a conference on polls and forecasts of the 2024 election. Most of the attendees were pollsters or academics, but I also watched presentations from Aaru and Electric Twin, two companies that do what is interchangeably called synthetic sampling, silicon sampling, or synthetic audiences. Stripped of startup jargon, that means they use large language models (LLMs) to simulate responses to public opinion polls by having AI agents take on the role of survey respondents. I had already heard of Aaru thanks to some articles with eye-catching headlines like “No people, no problem: AI chatbots predict elections better than humans” in the months leading up to Election Day. The founders were making some big, some might even say far-fetched claims, such as: “within two years, we will simulate the entire globe — from the way crops are grown in Ukraine to how that impacts production of oil in Iraq, trade through the strait of Malacca, and elections for the mayor of Baltimore.” When Semafor asked Aaru’s cofounders — Cameron Fink and Ned Koh — about my boss, they said “we respect all those who came before us.” Nate (as he so often does) shared his thoughts on Twitter: Fink and Koh were relatively good-natured about this back-and-forth when we spoke at Oxford. They even offered to mail me one of the t-shirts featuring Nate’s quote they apparently had made. I never took them up on the offer, which I now somewhat regret. These synthetic sampling companies fell off my radar for a while, but they do still exist. In fact, Aaru recently received a $1 billion valuation. Is what they’re doing anywhere close to the most important frontier in AI development? Not by a longshot, especially when Anthropic just developed a model so adept at exploiting software vulnerabilities that it’s only being released to 40 companies. Still, silicon sampling is increasingly finding its way into public polling. Axios reported in March that “a majority of people trust their own doctors and nurses” based on findings from Aaru — without mentioning that the “people” in that sentence were actually LLMs. Around the same time, the Public Sentiment Institute “boosted” their online sample of 373 real survey respondents with 114 AI agents.1 (Spoiler alert: even the co-founder of Electric Twin doesn’t think that’s a particularly defensible approach.) Polling companies like Qualtrics and Ipsos are also developing synthetic data panels. So, what should we make of these … “polls”? Let’s get one thing out of the way: whatever they are, they’re not polls in the way that term is usually defined. Subscribe You can’t replace polls with AI On one hand, using LLMs to essentially make up fake survey respondents sort of sounds like the dumbest idea ever, one that will at best imperfectly replicate real polls while introducing all sorts of biases. On the other hand, with LLMs improving at a remarkable, perhaps even alarming rate, maybe that means I’m a dinosaur at the ripe old age of 24 because I still want to rely on polls that talk to actual people. I’m not going to argue that synthetic samples are completely useless. In fact, as I’ll return to later, there is evidence that some techniques can replicate topline survey results quickly and cheaply. But the marketing from certain companies can be slightly optimistic. “No traditional poll will exist by the time the next general election occurs,” said Fink in 2024. We’re just 206 days away from the midterms, and based on the fact that I still have to collect a bunch of polls every day, I’d say he should have run that prediction by a sample of AI agents before the interview.2 To see why synthetic samples can’t replace polls, here’s a quick primer on how they work. The simplest version of these models involves taking a LLM (like ChatGPT or Claude), giving it a demographic profile (e.g., a white, college-educated woman who lives in Utah and makes $70k a year), and then asking it to respond to a survey question. You repeat that process a few thousand times using different demographic profiles and end up with a sample of synthetic survey responses. The actual models used by private companies are more sophisticated than this, usually because they incorporate more hypothesized demographic characteristics for each agent and provide them with extra information. Aaru, for example, feeds agents a diet of news and information they’d be likely to consume, while Electric Twin incorporates their customers’ proprietary data about the audience they’re trying to replicate. The way Ben Warner, the co-founder of Electric Twin, explained it to me was “we have a large amount of data on […] for instance, 5,000 people. Can we make an accurate prediction of how they would respond to another question?” Still, it should be obvious why synthetic samples can’t replace polls. Polling is fundamentally a data collection process. We might use surveys to make predictions by feeding them into election forecasts, but the main purpose of a poll isn’t prediction, it’s gathering new data about what people think and how they feel. Silicon sampling, on the other hand, produces no new data. It’s simply a model: you input LLM training data, demographic prompts, and a bunch of other information, and it spits out a prediction for what a poll would say. We love models here, but models aren’t polls. That difference is an important philosophical sticking point for most pollsters I talk to. “I think politics should stay away from [synthetic sampling], because we’re trying to […] represent the voice of the people,” said Natalie Jackson, a vice president at GQR Insights. Democratic pollster John Hagner told me: “I think I’m just incredibly skeptical of this idea. I don’t think it’s research. At that point, you’re asking the machine to tell you what you already believe.” Hagner has seen some presentations of early synthetic sampling experiments, but so far, “if it’s being used in a campaign, people are keeping it incredibly quiet.”3 But Eli, I hear you saying, aren’t polls themselves increasingly governed by modeling decisions? Indeed they are: pollsters’ choices on which sampling method to use, how to define their likely voter models, and how to weight their samples can and do lead to dramatic differences in the results they publish. Aaru even referenced these limitations in the methodology statement included with that maternal mortality “poll” — although I’m using the term “methodology statement” loosely here, because it doesn’t really explain how the model works at all. We can ignore the (frankly preposterous) implication that synthetic sampling isn’t subject to a separate set of biases. The important point is that there’s still a meaningful difference between using weighting and other statistical techniques on actual polling data and using a model to predict what a poll would say. The latter is far closer to election forecasts or techniques like MRP — potentially useful models, but not a replacement for polls.4 To be fair, other synthetic sampling companies are perfectly happy with the distinction between polls and models. Warner compared polling and synthetic sampling to different tools in a toolbox. “The mistake I think we make is we think that these new tools should either work in exactly the same way or somehow replace these old tools,” he said. “Rather than thinking of it as, okay, so we’ve always had the hammer, we’ve always had the screwdriver, now we’ve got a saw. But don’t use a saw to try [to] do the job of a hammer.” A quick comment from Nate Eli didn’t ask me for a comment — rather rude of him, don’t you think? But since I’m editing this story, I figured I’d add a few quick thoughts rather than putting words in his mouth. Beyond the frequently misleading marketing, what bothers me about the AI “poll” hype is that as AI tools make statistical inference cheaper and/or better (note that these are not synonyms) that actually increases the comparative value of collecting original data. You might be able to train a model to make a reasonable estimate of what some hard-to-reach poll respondent would say — say, a young Black man who voted for Trump. (Such a person checks a number of boxes for a voter who is usually hard to reach in surveys.) Indeed, this is closely related to what models like the Silver Bulletin forecast already do. They essentially smooth out the kinks in noisy survey data by making inferences based on past voting patterns or national polls or surveys of other states. But you don’t actually know what these voters think unless you’re reaching them directly. If there’s a shift in opinion among this subgroup, you’re not going to detect it. So if I were running a campaign, I’d invest more in going the extra mile to find a representative sample of those voters. And then I’d hire some smart quants — or Claude? — to figure out the implications for campaign strategy based on proprietary data that my competitors didn’t have access to. -Nate Silver Are these models any good? If synthetic surveys are just a new type of model, the next obvious question is whether the models are at least accurate. The answer very much depends on who you ask. On one end of the spectrum, you have the maximalist argument that synthetic sampling is more accurate than actual polls. “It’s an incredibly challenging problem to go to someone and say ‘hey, we’re going to be more accurate at predicting human behavior than you, even when you talk to your customers directly’,” Koh recently told CNBC. In his view, synthetic sampling isn’t a saw to polling’s hammer, it’s “magic.” There’s certainly evidence that synthetic samples can replicate certain survey toplines. But if Aaru does have any examples of their approach outperforming the polls, they’re keeping those to themselves.5 Aaru’s 2024 election model, for example, had Kamala Harris leading in Michigan, Nevada, Pennsylvania, and Wisconsin on November 4th. And although they’ve since taken down their forecast page, they gave Harris a 50.5 percent chance of winning the race on November 2nd.6 After the election, Fink told Semafor he was happy enough with those results because they were “within margin of error,” a term that is completely meaningless when applied to a “sample” of AI agents. And of course, Aaru says their models have improved since 2024, so supposedly now they’d be more accurate than the polls? Still, their stronger argument is on cost: “We are significantly faster and cheaper than traditional polling, and still more accurate,” said Fink. The first two claims are undeniably true, but the third brings us to the opposite end of the spectrum. Both Jackson and Hagner are skeptical that these models are reliable for anything beyond replicating common survey toplines. “I just […] don’t think the machines are what we want when we’re looking for nuanced views. My example on this is people in Arizona and Nevada in 2024 who voted for Trump and voted for expanding abortion in their states on ballot initiatives,” said Jackson. Hagner identified another issue. Maybe the synthetic respondents, like sycophantic LLMs, are inhuman in one important way: they’re too nice. “The reports that have come through at the meetings that I’ve been at are that the early experiments on this, they cannot get respondents to be as racist or sexist or, frankly, as negative as human respondents,” he said. Academic research mostly agrees on this point. While there are some papers that show promising results when using LLMs to replicate polling data, most show that LLMs suffer from various quirks like producing too few “don’t know” responses and can seriously overpredict the favorability of politicians like Donald Trump and Kamala Harris. They also seem to struggle with too little variation between demographic subgroups, so the difference in predicted opinion between Democrats and Republicans, for example, is too small. When I asked Warner about these studies, his response to these papers was that just because academics can’t get synthetic sampling to work doesn’t mean that the technique doesn’t work in general. “Actually, the argument is, okay, yours does not [work]. That does not mean […] for this complex set of machinery, which uses a lot of investment, a lot of time, a lot of money, you can’t get it to work.” Cards on the table, I’m somewhat sympathetic to this argument because academics aren’t exactly great at making election forecasts. Usually, the people with skin in the game are the most accurate. Warner’s argument is that the approach Electric Twin takes — which includes, for example, making multiple predictions for each synthetic respondent using different models and prompts and subsequently averaging those to get a final prediction in a sort of ensemble forecast — produces better results than the simpler academic models. Warner shared a comparison between his method and the method from a recent academic paper with me, and Electric Twin was indeed able to get more accurate replication. But even still, he acknowledged that synthetic sampling “is not a crystal ball.” “If you asked me, do I think using other data sources will be more accurate than asking somebody who they will vote for, I would probably say no. But if you asked me ‘would your system be useful for our turnout modeling today?’ I would say yes.” For better or worse, it looks like the method is already getting more popular in the market research world. Most of the clients Aaru touts these days are businesses like EY and McDonald’s. And AI will probably begin to pop up in other parts of the political polling process. Pollsters are already using it to code open-ended survey responses, and some firms, like YouGov, are testing using LLMs to ask survey respondents questions. More worryingly, one danger to actual polls is that AI agents can be used to infiltrate online surveys. Most online polls use various checks to prevent that from happening, but there’s conflicting evidence on how effective those filters are and how prevalent AI agents currently are in online panels. If those agents ever become impossible to detect, it might spell the end of online polling, but the solution isn’t to replace all of your respondents with ChatGPT. Silver Bulletin is a reader-supported publication. To receive new posts and support our work, consider becoming a subscriber. Subscribe 1 That particular poll obviously doesn’t meet Silver Bulletin standards for aggregation. But we exclude all Public Sentiment Institute polls from our averages because we classify them as an amateur polling firm. 2 You could argue that Fink meant the next presidential election, but (a) I’m also confident we’ll still have real polls in 2028 and (b) in that case he should have asked an LLM to define “general election.” 3 Quick caveat: that’s reporting from a Democratic pollster. It’s possible that Republicans are more willing to use AI in political campaigns. 4 Indeed, Silver Bulletin does not include “polls” produced by MRP in our forecasts or averages, and we think it’s extremely misleading when their practitioners describe them in a way that suggests original data had been collected among a large number of states or Congressional districts. 5 A recent report from Aaru and EY did show two examples of a synthetic estimate being closer than a survey to a real-world benchmark — but I’d take those findings with a grain of salt because the report reads more like an ad and didn’t involve any sort of prediction being made ahead of time. 6 For comparison, our odds for Harris on the same day were 48.2 percent.
Russian Memorial human rights NGO labeled 'extremist'
Russia's Supreme Court has labeled the human rights organization Memorial as "extremist." Experts say this represents a new level of repression in Russia.
The Climate Issue: The blue marble, then and now
Our planetary-affairs editor, Oliver Morton, looks at the connection between space travel and climate science
'I Saw How It Was Getting Damaged': Ex-Bethesda Exec Goes to Town on Xbox's Mistreatment
"That shouldn't be a surprise to you"
Photos: Makeup tests, shoulder cuff health and student presentations
Darryna Guy-Williams, a bioengineering and bio chemistry major, sprays Jenna King, class TA and postdoctorate candidate in chemistry and chemical biology, with a facial hydrating mist in Leila Deravi’s cosmetic chemistry class.
Why the World’s Biggest TikTok Star’s $975 Million Business Deal is Suddenly Falling Apart
Khaby Lame’s landmark merger drew in traders and creator economy optimists alike, but the stock has since plunged more than 90 percent and major brokerages are restricting trading.
Trump administration admits a glaring error in its accusations about New York health care fraud
NEW YORK (AP) — President Donald Trump's administration this week acknowledged it made a significant error in figures it used to help justify a fraud probe into New York’s Medicaid program, a glaring mistake that undercuts a federal campaign to tackle waste, mostly in Democratic-led states.
The US in Brief: The Pentagon’s battle against free speech
Our daily political update, featuring the stories that matter
Ohio economists: Water-quality bonds good for public health, uncertain about other benefits
Economists agreed that water-quality bonds would benefit Ohioans. But they were less certain that they would grow the state economy.
Making Friends In Politics Is Possible
I knew getting involved in politics was a great way to make enemies, but it never occurred to me until I saw it in action that it can also be a way to make friends. I’m still not very involve…
AI micro-dramas are shaking up Chinese entertainment
Regulators threaten to spoil the fun
AI-generated micro-dramas are shaking up entertainment in China
Regulators threaten to spoil the fun
The US in Brief: Jaw-jaw, not war-war?
Our daily political update, featuring the stories that matter
The Future of Colorado condo development
By Peter LiFari For nearly two decades, the story of condominium construction in Colorado has been mistold, misunderstood and misrepresented. This outcome reflects the cumulative effect of more than two decades of Colorado policy, beginning with the Construction Defect Action Reform Act of 2001 and evolving through subsequent legislative changes, that have reshaped the risk landscape governing […]
Nancy Guthrie’s kidnappers may have made fatal health mistake: Detective
“It just scares me to think that somebody maybe took Miss Guthrie, and she had some sort of medical event.”
Jacksonville Bold for 4.8.26: Learning from mistakes
Republicans regroup for Jacksonville Mayor race as campaign season intensifies.
Nancy Guthrie Kidnappers Likely Made This Fatal Mistake, Says Investigator
The 84-year-old mom of Savannah Guthrie has been missing for ten weeks.
Nancy Guthrie Kidnappers Likely Made This Fatal Mistake, Says Investigator
The 84-year-old mom of Savannah Guthrie has been missing for ten weeks.
People visiting Japan are too scared of breaking rules, says local
As a resident of Japan, I've seen tourists to my country scared to break rules or make etiquette mistakes. But respect matters more than perfection.
The US in Brief: What’s next?
The latest news in US politics, with coverage of Donald Trump’s second term.
What smart people are saying about Trump's ceasefire deal with Iran
Economists, foreign policy experts, and politicians weigh in on Donald Trump's last-minute ceasefire deal with Iran.
Viewpoint: FP10 should remain open, excellent and secure
Europe is right to worry about research security. Foreign interference is real. So is the misuse of sensitive knowledge. Universities working in sensitive areas such as AI, quantum, semiconductors and advanced materials cannot afford to ignore the geopolitical environment around them.But Europe is in danger of drawing the wrong conclusion, one that would increase, not reduce, its strategic vulnerabilities.If the EU applies an “EU preference” logic to the next Framework Programme for research and innovation (FP10), it will weaken one of its greatest strategic assets: a research system built on openness, excellence and international collaboration. In the name of security, Europe could end up advancing science more slowly, innovating less effectively and making itself less attractive to global talent and partners in advanced science and technology.That would be a serious mistake.The EU preference included in Article 10 of the proposed European Competitiveness Fund may make sense in some industrial or deployment contexts, but is far harder to justify when applied to collaborative research in advanced science and technology, where excellence and global connectivity are crucial. Research is not procurement. It does not thrive when isolated. It advances through access to the best ideas, the best people, the best facilities and the best data, wherever they are found. This is why FP10 should not simply absorb industrial policy logic by default. Doing so would risk importing instruments that are ill-suited to research and that ultimately increase vulnerability rather than resilience.For universities of science and technology, the issue is especially acute. These institutions sit at the frontier of knowledge and technology creation in the very fields Europe sees as critical for its future competitiveness and resilience. They know better than most that security risks are real. But they also know that openness and security are not opposites. In practice, they have to be made to work together. Research security is not about closing doors. It is about building resilience through collective responsibility, trust, proportionate safeguards and better governance. That is a fundamentally different approach from default restriction.The risks of non-collaborationThe real danger in the current debate is that Europe talks too much about the risks of collaboration and not enough about the risks of non-collaboration. Those risks are strategic too. Cut researchers off from international networks and Europe loses access to knowledge, data, infrastructures and talent. Slow down global collaboration in critical technologies and Europe slows down its own progress on cutting-edge science. Make participation rules unpredictable and Europe weakens its attractiveness to researchers and investors alike.In other words, over-restricting collaboration will create strategic vulnerabilities, particularly in advanced science and technologies.This point is often underestimated in policy discussions that frame openness mainly as a value to be defended. Europe’s scientific strength, technological development and long-term competitiveness depend on remaining connected to the best research ecosystems in the world. For a continent that wants to accelerate deep tech, strengthen industrial capacity and reduce strategic dependencies, weakening the excellence base would be self-defeating. That does not mean ignoring risk. Quite the opposite. Universities are already strengthening due diligence, internal governance, awareness-raising, screening procedures and compliance support. Across Europe, institutions are working to manage risks linked to foreign interference, sanctions, export controls and the misuse of research. The right response is not complacency. It is better risk management. But that means targeted, evidence-based and proportionate measures, not blunt top-down restrictions with wide spillover effects. In the daily reality of universities, openness and restriction often coexist. The real challenge is how to apply safeguards without sliding into securitisation that damages academic freedom, open science and international cooperation. Restrict partners, not nationsMore restrictive treatment of collaborators from European Economic Area members Iceland, Liechtenstein and Norway, or from Switzerland and the UK, is strategically very difficult to justify. These are deeply embedded partners in Europe’s research and innovation ecosystem. Beyond that, Europe should still resist default exclusion and instead rely on case-by-case assessment. The relevant question is not nationality in the abstract. It is the actual risk profile of a collaboration: the institution, the affiliations, the field, the knowledge involved, the intended use and the safeguards available. Screening should focus on institutions and affiliations rather than nationality alone. Related articlesParliament ‘misreads’ link between Horizon Europe and ECF, Commission official saysFrustration grows at governance gap in next Horizon EuropeIf EU preference logic is imported into research without careful calibration, the likely result is a chilling effect. Researchers may step back from legitimate and beneficial collaboration, even where risks are manageable. International partners may see Europe as less predictable. And Europe may end up undermining the very excellence base it needs to compete, scale up and lead in strategic technologies, thereby increasing its dependence on others.FP10 should do the opposite. It should remain open by default, with restrictions used only exceptionally, and then on the basis of clear evidence. It should integrate research security into open science and research integrity frameworks, not set them against each other. And it should recognise that academic freedom, scientific excellence and responsible international collaboration are not obstacles to Europe’s security, they are part of it. Europe needs stronger tools to protect itself. But it also needs the discipline to avoid measures that would weaken it and increase its strategic vulnerabilities in the process. In FP10, that means staying open, excellent and secure.Louise Drogoul is senior advisor for innovation and sustainability at the Cesaer university association.
How Trump Took the U.S. to War With Iran
In a series of Situation Room meetings, President Trump weighed his instincts against the deep concerns of his vice president and a pessimistic intelligence assessment. Here’s the inside story of how he made the fateful decision.
More business openings than closings in Oregon in last decade, despite dour economic stigma
Far more businesses have opened than closed in Oregon during the last decade according to analysis from an Oregon State University economist.
The world’s “oldest octopus” was never an octopus
A famous “oldest octopus” fossil has been exposed as a case of mistaken identity. Advanced imaging revealed hidden teeth showing it was actually related to a nautilus, not an octopus. The confusion came from decay that altered its shape before fossilization. This discovery rewrites part of evolutionary history, pushing the true origin of octopuses much later in time.
The US in Brief: A(nother) ultimatum
The latest news in US politics, with coverage of Donald Trump’s second term.
NYC health department has questions about the floating pool set for the East River
The pool’s creators are adding an emergency chlorine dosing system as they await health department approvals.
Trump holds fast to Tuesday deadline, threatening Iran's bridges and power plants
Trump said he was optimistic about the prospects of working with a newly installed Iranian government. But Tehran dismissed the diplomatic overtures as “unrealistic."
Mass. business confidence slips back into pessimistic territory
Amid ongoing conflict in the Middle East, Massachusetts’ business confidence returned to pessimistic territory in March.
NYC's top economic agency still leaderless under Mamdani. Business leaders are worried.
Mayor Zohran Mamdani is approaching 100 days in office. He still hasn't found someone to run the city's Economic Development Corporation.
The biggest mistake HR leaders make when pitching new benefits to their CFO
HR leaders who pitch wellness programs by leading with employee happiness are making a costly mistake. The secret? Crunch the numbers.
New Study Reveals Hidden Ocean Chemistry
Science team led by Mariana Bif (right) prepare to launch a Biogeochemical-Argo (BCG Argo) float.
A Look At Sphere Entertainment (SPHR) Valuation After Guggenheim’s Bullish Update And Las Vegas Event Momentum
What triggered Sphere Entertainment’s latest move? Sphere Entertainment (SPHR) drew fresh attention after Guggenheim responded to strong quarterly results and the Sphere Experience’s Wizard of Oz performance, alongside a packed Las Vegas event schedule, with more optimistic revenue and operating income assumptions. See our latest analysis for Sphere Entertainment. Sphere Entertainment’s recent 6.18% 1 day share price return and 19.15% 7 day share price return, taking the stock to US$127.22,...
New study reveals hidden ocean chemistry
Findings reveal that nitrogen cycling in marine oxygen-deficient zones is far more dynamic than previously thought
New study reveals hidden ocean chemistry
Scientists have found a new way to detect subtle chemical signatures in seawater—revealing previously invisible details about the ocean’s chemistry from data continuously collected by thousands of autonomous robotic floats drifting across the seas.
Apple’s Update Mistake—Hundreds Of Millions Of iPhones Affected
Why you need to change your iPhone settings now — don't leave this until it's too late.
When hospitals become corporations, what happens to patient safety?
In many medical malpractice cases, the public imagines a careless doctor making a single catastrophic mistake. In reality, serious medical errors are often the result of systemic pressures that pus…
What’s the most attractive genre of entertainment?
Millions of people globally tune in to “beauTV”
Mars dust storms are sparking electricity and rewriting the planet’s chemistry
Mars may look like a quiet, dusty world, but it’s actually buzzing with hidden electrical activity. Powerful dust storms and swirling dust devils generate static electricity strong enough to spark faint glowing discharges across the planet, triggering chemical reactions that reshape its surface and atmosphere. Scientists have now shown that these tiny lightning-like events can create a surprising mix of chemicals—including chlorine compounds and carbonates—and even leave behind distinct isotopic “fingerprints.”