SCROLL NEWS / DISCOVERY
Search headlines
Find the stories shaping the conversation.
Results for " going " 3,942 found 🔀 AI-powered Shuffle
“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.
All eyes are going to be on Justin Bieber at Coachella
Justin Bieber’s headlining set at Coachella on Saturday will be important moment in his slow-walked return to the stage following a turbulent few years marked by health struggles and cancelled tours.
Historic live-donor transplant at University Health saves 7 lives
Fourteen people made history at University Health this year, undergoing surgeries that saved seven lives.
Memorial Health Foundation fashion show raises over $142K
The Memorial Health Foundation 21st annual Fashion Show, this year with a Masquerade Ball theme, drew a large crowd and raised more than $142,000, with proceeds supporting local cancer patients and related health initiatives. The annual fundraiser supports the foundation’s cancer patient emergency fund and breast health initiatives, helping individuals undergoing treatment with expenses such […]
Cowboys should consider this offensive player in 1st round
Is there any way you would be okay with Dallas going offense in the first round?
Teddi Mellencamp reveals she's dating an older man who’s been 'pretty good' amid her ongoing cancer battle
Teddi Mellencamp reveals she's dating an older man who has been 'pretty good' amid her ongoing cancer battle.
Chris Brown and Usher are going on tour together: 'It's time'
Chris Brown and Usher going on tour together, the pair announced on Instagram.
Hundreds of chimpanzees are killing each other in a 'civil war' — scientists captured the 'lethal violence' for the first time
They're going apes—t.
WHO Says Iran Health System Is in Crisis, Commits $800,000 Emergency Funds
The World Health Organization released $800,000 from its emergency contingency fund to support the health response in Iran. Hanan Balkhy, a regional director for the United Nations agency, welcomed the current cease-fire but warned the health crisis in Iran was ongoing. “Humanitarian and health nee
YouTube Premium is getting pricier
The standard plan is going up by $2 a month in the US.
YouTube Premium is getting a US price hike of up to $4/month
YouTube Premium is getting a price hike in the US, with all plans going up in cost by up to $4 effective over the next few months.
Deadly Philly parking garage collapse leaves Grays Ferry shopping center, businesses closed
Grays Ferry residents and businesses are dealing with ongoing closures and disruptions after a partial parking garage collapse, with demolition set to begin Friday morning.
‘I don’t see any new faces’: Rochester Road construction changes daily life for small businesses in Troy
Some businesses along Rochester Road in Troy say ongoing roadwork is cutting into their bottom line.
Pulaski County business raises funds for Taylor Regional staff amid delayed paychecks
HAWKINSVILLE, Ga. — Concerns continue to grow surrounding Taylor Regional Hospital, as employees deal with delayed paychecks and ongoing uncertainty. Now, a loc
City of Bryan provides health update on City Manager Andrew Nelson
The City of Bryan announced Thursday that City Manager Andrew Nelson is undergoing treatment for a mass found in his brain.
As Trump throws lifeline to coal plants, critics warn of higher costs and health risks
The United States is undergoing a dramatic shift in energy policy as President Donald Trump wields government's sweeping powers to benefit coal and suppress cleaner alternatives. It could lead to more expensive electricity and dirtier air.
America’s Furniture Stores Struggle to Survive a Frozen Housing Market
Retailers are going bankrupt and liquidating as record-low housing turnover leaves fewer customers looking to furnish homes.
3L Rachel Sadler Charts Path in Brain Health, Technology, and Policy Through UC Law SF’s Health Law Concentration - UC Law San Francisco (Formerly UC Hastings)
3L Rachel Sadler studies health law and policy at UC Law SF while researching the legal implications of neurotechnology and other emerging tools that may affect human cognition. 3L student Rachel Sadler shares how UC Law SF’s Health Law Concentration prepared her for a career focused on the intersection of brain health, government policy, and technology. Gained a detailed understanding of how U.S. health care is regulated and where policy reform is needed through coursework and research. Interned with state and federal agencies that regulate health care services and worked with a major health care provider’s bioethics department. Plans to pursue a career focused on protecting cognitive liberty and mental privacy as new technologies reshape healthcare. 3L Rachel Sadler: The Health Law Concentration has given me fluency in one of the most complex regulated industries and the tools to become an effective advocate for better health policy. Before coming to law school, I spent a decade teaching and working with teachers and administrators on high school programming for multilingual learners. Healthcare kept emerging as the most powerful driver of student outcomes. I wanted to understand how to make effective policy change in this area and how to use existing law to defend people’s health rights in California. Through the concentration’s coursework, I gained a sophisticated understanding of how U.S. healthcare is regulated. I also learned where the pressure points for reform exist. The courses gave me space to develop what became my central research interest: how law and policy can be built to protect human cognition and healthy brain development, particularly regarding new technologies such as generative AI and neurotechnology. The health law faculty were instrumental in that process, genuinely invested in helping me and other students define and pursue our own intellectual vision. During law school, I interned at the General Counsel’s Office for Region IX of the U.S. Department of Health and Human Services. I also externed at the California Department of Justice’s Health, Education and Welfare Division. Additionally, I externed at Kaiser Permanente Northern California’s Bioethics Department. Alongside those placements, I conducted extensive independent research on mental privacy and emerging neurotechnology, working closely with Professors Emily Murphy and George Horvath. That research became the throughline of my law school experience and the foundation of my current policy focus. I must also credit Nicole Ozer of the Center for Constitutional Democracy (CCD), whose deep knowledge of California privacy law and the public-interest privacy landscape deeply informed my work. I currently serve a research assistant with the CCD, where I spearhead the development of K-12 educational resources in partnership with law faculty and students. I’ve found UC Law SF to be an ideal place for second-career, nontraditional students. There is a tremendous diversity of life and career experience here, and I felt fully welcomed into the community, despite returning as a student at a different stage of life. I came to law school wanting to promote equity in California’s healthcare system. I am leaving with a sharper and more specific ambition: to build a policy career focused on protecting cognitive liberty and mental privacy from technological intrusion. This is both an urgent and underserved area of law, and I want to be part of defining this landscape going forward. Whether as a practicing attorney or a policy advisor, I plan to work on these issues after law school. I strongly recommend UC Law SF and the Health Law Concentration. When I arrived in law school, I worried the concentration would be narrow and that my non-clinical background would be a disadvantage. The opposite was true. The health law faculty thinks about health expansively, encompassing social determinants of health, structural inequity, and emerging technologies alongside the regulatory mechanics of the industry. The faculty are accessible, intellectually generous, and genuinely invested in students’ development. I found my research vision here, and I deeply appreciate the concentration’s role in this. The Evidence of Success series highlights UC Law SF students as they share how the College’s opportunities equip them with the experience, skills, and confidence to excel in the legal profession.
‘Jeopardy!’ Contestant Who Almost Beat Jamie Ding Speaks Out
Dominex Kovacs shared how he knew going all in on the Daily Doubles was the key to beating him.
What’s really going on at UT Austin? Inside UT goes behind the scenes
This free newsletter will deliver reporting from the Forty Acres and include exclusive, behind-the-scenes details she won’t share elsewhere.
The best sleepwear may be no sleepwear. Here’s what experts say
Sleeping too hot can ruin rest. Experts explain how cooler bedding, breathable fabrics or going naked may improve sleep and more.
WTOP Elevates Local Sports Coverage with Expanding Focus on Video and Analysis
FOR IMMEDIATE RELEASE Washington, D.C. (April 9, 2026) —WTOP announced the promotion of Rob Woodfork to Senior Sports Analyst, marking a significant step forward in the organization’s ongoing…
Pakistan’s Last-Minute Diplomacy Secures Iran War Ceasefire
Pakistan warned U.S. officials that ongoing Israeli operations risked collapsing the entire diplomatic process.
iPod revival: Secondhand sales surge for Apple's digital musical player
The iPod is undergoing a revival, with secondhand sales surging. Young people in particular are drawn to its retro look and the focused music experience it offers, free from smartphone distractions.
Injury-plagued Lakers look to regroup vs. Warriors
The foundation is crumbling fast around the Los Angeles Lakers, who looked like championship contenders after going 15-2 last month.
Golden Entertainment takes next step toward going private
If the Gaming Commission approves later this month, the owner of the Strat, Arizona Charlies and the PT’s Pub chain will transition from being a publicly traded company.
Apple seeks records from Samsung in South Korea in antitrust case
Apple has moved to obtain internal Samsung data from South Korea in its ongoing antitrust fight with the DoJ over alleged monopoly practices.
Ishbia family taking swing at youth sports facility for vacant Office Depot
(Editor’s Note: Alex Harrison Perry reported this story for Evanston RoundTable, a neighboring independent newsroom. It was shared with The Record as part of an ongoing collaborative effort.) The vacant Office Depot building at 2722 Green Bay Road — just a block south of Wilmette — could become a year-round site for youth sports, as a North Shore family […]
Nevada regulators approve Golden Entertainment going private
The CEO of Golden Entertainment told Nevada regulators that a lagging stock price and investors not properly valuing the casino operator caused them to take the company private. The Nevada Gamin...
I On Politics
DUTCH KILLS CIVIC ASSN MEETING APRIL 9: The next general meeting of the Dutch Kills Civic Association will be held on Thursday, April 9, at 7:00 pm at the Growing Up Green school, 39-27 28th Street. Joseph Ottomanelli from the NYC Sanitation Dept will be the guest speaker discussing the bin mandate going into effect
If Vermont is going to scale up primary care, health advocates say this one funding priority is crucial
The House-approved budget for next fiscal year gives scholarship and loan repayment programs more money than the governor originally proposed, but many worry it will not be enough.
Sony Pictures Entertainment to cut hundreds of jobs in TV and film
The job cuts at Sony Pictures come amid a deep slump and existential crisis in Hollywood, which has faced setbacks, two labor strikes and the ongoing collapse of cable TV.
Brad Stevens embraces Celtics role, shuts down college coaching rumors
Boston Celtics President of Basketball Operations Brad Stevens made it clear that he's not going anywhere, despite the rumors.
Quantum computers keep losing data. This breakthrough finally tracks it
Quantum computers struggle with a major flaw: their information vanishes unpredictably. Scientists have now created a new method that can measure this loss over 100 times faster than before. By tracking changes in near real time, researchers can finally see what’s going wrong inside these systems. This could be a big step toward making quantum computers stable and practical.
Boones Mill business owners claim landlord refused to fix issues with buildings, resulting in their businesses failing
Business owners in Boones Mill say ongoing maintenance problems at a commercial property forced them to shut their doors, raising concerns about landlord responsibilities and tenant protections.
Last Call for 4.7.26 – A prime-time read of what’s going down in Florida
A digest of the day's politics and policy while the bartender refreshes your drink.
Sporting CP 0 Arsenal 1 – Gyokeres’ timid return, Raya’s superb saves and Arteta’s savvy subs
A late goal from Kai Havertz secured a valuable advantage going into the second leg of this Champions League quarter final
The First Reviews Are Going Live For Starfield's Terran Armada Expansion
Is it worth grabbing the new DLC?
Willson Contreras lashes out at Brewers after 24th career HBP: 'I'm going to take one of them out'
Contreras was hit by a Brandon Woodruff pitch in the third inning Monday night
Willson Contreras after being hit by Brewers: 'Next time they hit me again, I'm going to take one of them out'
If the Lakers are going to survive their recent injuries, they'll need one last miracle from LeBron James
Can James turn back the clock enough to buy his teammates recovery time?
MAGA Is Winning Its War Against U.S. Science
Source: NBER With all the other terrible news right now, you may not have noticed that Donald Trump is in the process of killing American science. OK, that’s an exaggeration — but not that much of an exaggeration. The Trump administration’s latest budget proposal calls for a gigantic increase in military spending combined with severe cuts to social programs. But as the chart above shows, it also calls for debilitating reductions in research funding. Furthermore, Trump appointees have already been strangling science by sharply reducing the rate at which research grants are approved. Here, for example, is the number of new grants approved by the National Science Foundation: Source: Nature Large numbers of existing grants have also been frozen or terminated, especially in the study of infectious diseases. Add to this a sharp drop in visas issued to foreign students, who often play a direct role in research and who help support academic departments that do research: Put all of this together, and much U.S. scientific research is set to come to a screeching halt — not a few years from now, but over the course of the next year or two. This new assault on U.S. science is taking place at a time when the role of American science in the world has already been greatly eroded. The chart below, based on research recently reported by the National Bureau of Economic Research, measures national strength in science by the share of publications in highly ranked journals. In the 1990s the United States had more such publications than the rest of the world combined. Since then we’ve dropped into third place — well behind China and slightly below the European Union. And this was before the Trump administration’s attack on science had time to take its full effect. Source Some erosion in U.S. scientific preeminence was inevitable given China’s growing sophistication and wealth. But we’ve also fallen behind Europe, even though everyone says that Europe is lagging economically and technologically. Claims about Europe’s underperformance are, in fact, dubious if one looks hard at the data. But it’s still striking to see America lagging. What’s going on? There are presumably multiple factors behind America’s scientific lag. But even before Trump II the growing hostility of the U.S. right to science surely had some negative effect. And since the rise of MAGA G.O.P. attitudes toward science in general have become overwhelmingly hostile. This is true even for the Republican rank and file: And while I haven’t been able to find good survey data, it’s obvious that the anti-science turn has been even more pronounced — and began earlier — among the Republican political elite. Chris Mooney published The Republican War on Science in 2005, and even then was describing a longstanding trend. Why have Republicans turned so anti-science? Part of the answer is that they believe that scientists don’t support them. And they’re right! A study of who scientists give money to shows that only a small percentage gave money to Republicans even 20 years ago, and that almost none of them donate to Republicans now: Social scientists have always been strongly pro-Democratic, while there used to be a significant number of physical, “hard” scientists supporting the GOP. But these days physicists are almost as uniformly Democratic, or at least non-Republican, as sociologists. Why are there almost no Republican scientists? It’s not a mystery. GOP political orthodoxy includes positions that are at odds with the scientific consensus on multiple issues, ranging from the validity of the theory of evolution, to the reality of climate change, to the efficacy and safety of vaccines. In each case the scientific consensus is solidly grounded in evidence. But even before the rise of MAGA the U.S. right was increasingly hostile to evidence-based policymaking — especially, of course, where the evidence is unfavorable to fossil fuel interests or quack medicine, both financial mainstays of right-wing politics. So scientists don’t support Republicans, and the feeling is mutual. Today’s Republican Party doesn’t like science or scientists. It doesn’t like having its preconceived views challenged by appeals to evidence. It knows that very few scientists are on its side electorally. In general, it sees scientific research as a threat to its grasp on political power. Add in MAGA’s combination of rabid anti-intellectualism and allergy to any hint of criticism, and one has the makings of a drastic anti-science turn in policy. “Ignorance is strength” might was well be an official MAGA motto. And as I said, we aren’t talking about something that will happen over the course of multiple years: The U.S. scientific enterprise is threatened with severe damage, even collapse, over just the next year. There are many reasons to find this prospect horrifying: Think of all the beneficial advances, affecting almost every part of life, that won’t happen because U.S. science — still crucial to the world — has been eviscerated. But think, also, of America’s international standing. Can a nation that has forfeited its role as a leader, or even a contender, in global science, still be a Great Power? No. MUSICAL CODA
Yet again, when rights are in conflict
I admit, it is going to take quite a bit of self-control to not have this entire column end up being yet another screed about a certain president who continues to shock the world in his race to the bottom of intellectual discourse. Like many, I suspect, when I first read the social media post […]
Think Twice Before Asking ChatGPT About Your Health
Obsessively Googling symptoms is bad. Going down an AI rabbit hole may be worse.
A Redistricting War in Florida, Georgia’s Runoff and the Latest Politics News
As deadlines approach in the next two weeks, neither is going quite according to the partisan plan.
iPhone Fold Enters Trial Production Phase Ahead of 2026 Launch
Apple partner Foxconn has started trial production on the iPhone Fold, according to Chinese leaker Instant Digital. Trial production comes before mass production, which Apple plans to start in July as long as no issues come up during the earlier testing stage. So far, Apple remains on track to launch the iPhone Fold in 2026, though the latest rumors suggest that it's not going to come out in September. Instead, it could launch sometime after the iPhone 18 Pro, debuting as late as December.
Expanding Digital Horizons: A Practitioner's Guide to the Newly Expanded IRS Business Tax Account — Current Federal Tax Developments
On April 6, 2026, the Internal Revenue Service released News Release IR-2026-46, marking a significant milestone in the agency's ongoing service improvement efforts. For CPAs and EAs navigating the daily complexities of tax administration, the newly expanded Business Tax Account (BTA) platform promi
Mass. business confidence slips back into pessimistic territory
Amid ongoing conflict in the Middle East, Massachusetts’ business confidence returned to pessimistic territory in March.
'Your doctor, your price': Byron Donalds rolls out health care plan
'We're going to make sure that there is transparency in health care.'
Terry Crews' Wife, Rebecca, Reveals Parkinson's Diagnosis, Had New Treatment: EXCLUSIVE
She's sharing her diagnosis publicly for the first time after undergoing an exciting new treatment procedure.
Trump endorses Republican Steve Hilton in California governor’s race
President Donald Trump has thrown his support behind Republican Steve Hilton in the California governor’s race, potentially shaking up a crowded race to replace outgoing Gov. Gavin Newsom.
The Business of KAWS: What Data and a Museum Show Reveal About His Market
It's time for another look at the global craze for KAWS and what keeps it going strong as his museum shows break new ground.
"What's going on?!!": World of Warcraft just created a truly insane, epic moment in gaming history — I had no idea Blizzard was still capable.
I have no idea how Blizzard pulled this off, but during World of Warcraft's Mythic raiding "Race to World First" esports event, U.S. team Liquid discovered live to thousands that the final boss had an insane, secret final phase.
Return to Sport Rates Following Total Knee Arthroplasty
A recent study shows that understanding timelines and key factors could help clinicians give tailored advice and set realistic expectations for their patients undergoing TKA.
Connecticut bars getting business boost from UConn basketball tournament runs
The UConn men’s basketball team run to the national championship is also bringing a lot of customers to local bars across Connecticut for watch parties. “We had the audio running and everything and as soon as the final buzzer hit, everyone was going crazy. We had a nice little UConn chant going at the end,” Yash Ramchandani, whose family owns…
Spokane Regional Health District to mark Public Health Week with events
Spokane Regional Health District (SRHD) is celebrating National Public Health Week from April 6-12, 2026, by highlighting the importance of coordinated public health efforts and showcasing ongoing initiatives to improve
Story of the birth of ESPN will be showcased in a documentary on Monday night
ESPN is going back to the beginning when it debuts a 90-minute documentary about its creation. “Sports Heaven: The Birth of ESPN” will premieres at 8:30 p.m. EDT on Monday.
Close Up: Highlighting women in history and Iowa politics
The contributions of female pioneers in Iowa's political history and their ongoing impact, and a candid conversation with CBS News Correspondent Norah O'Donnell
One Georgia football transfer remains in the portal
Fourteen of Georgia's 15 outgoing transfers found new schools with 11 of those players landing at Power Four destinations. One remains in the portal
Cowboys draft trade up rumors center on pass rush
Do you think the Cowboys are going to move around in the first round?
Golden Knights make the Pacific Division much more interesting
When the Vegas Golden Knights hired John Tortorella amidst a challenging stretch of play, many fans and pundits expressed curiosity and a dash of skepticism about if the change was going to awaken the sleeping giant.
NASA Disappoints Artemis II Astronauts By Cutting Off Chappell Roan Wakeup Song
This is a lot cooler than Katy Perry going into space: NASA played Chappell Roan’s “Pink Pony Club” to wake up the astronauts on Artemis II this morning. However, mission control abruptly cut the audio after Roan sang “heels.” “We were all eagerly awaiting the chorus,” Commander Reid Wiseman joked. Artemis II launched on Wednesday…
This $8 Harbor Freight Gadget Should Be The First Thing You Pack For Hotel Stays
There are certain things you shouldn't leave home without when going on a trip. For under $10, bringing this Harbor Freight gadget is a no-brainer.
Red Sox Anthony will get over early sophomore slump assumptions
The soon-to-be 22-year-old rising star is going to be just fine.
Tarrytown spa undergoing repairs after SUV crashed into business
Repairs are underway at a Tarrytown spa after an SUV crashed into the business. The business hopes to reopen in early May.
Colorado Supreme Court intervenes in 4 ongoing cases
The Colorado Supreme Court recently moved to intervene in four ongoing cases, including two criminal prosecutions, a civil lawsuit, and a child welfare proceeding. At least four of the court’s seven members must agree to hear a case outside of the normal appellate process. The issues raised before the justices include the ability of prosecutors […]
First Jam Kuradoberi match footage released for Guilty Gear Strive Version 2.00
Guilty Gear Strive is less than a week away from it's huge Version 2.00 update, and now we can get a taste of what that's going to look like with the newest character.Arc System Works and ...
New Apple TV Waiting for Siri: Here's What's Coming When It Launches
We're long overdue for an Apple TV update, and there have been rumors about an imminent refresh since late last year. It's now sounding like we're not going to get a new version for several months because of Siri delays. If you're holding out for a new model, here's a recap on what to expect when it eventually comes out so you can decide whether to continue to wait, or buy now. Design Apple TV design updates are few and far between, and that's not changing in 2026.
Here's Why Starfield Will Be Best on PS5 When It Finally Releases Next Week
Bethesda going all-out
Gemma 4 and what makes an open model succeed
Having written a lot of model release blog posts, there’s something much harder about reviewing open models when they drop relative to closed models, especially in 2026. In recent years, there were so few open models, so when Llama 3 was released most people were still doing research on Llama 2 and super happy to get an update. When Qwen 3 was released, the Llama 4 fiasco had just gone down, and a whole research community was emerging to study RL on Qwen 2.5 — it was a no brainer to upgrade. Today, when an open model releases, it’s competing with Qwen 3.5, Kimi K2.5, GLM 5, MiniMax M2.5, GPT-OSS, Arcee Large, Nemotron 3, Olmo 3, and others. The space is populated, but still feels full of hidden opportunity. The potential of open models feels like a dark matter, a potential we know is huge, but few clear recipes and examples for how to unlock it are out there. Agentic AI, OpenClaw, and everything brewing in that space is going to spur mass experimentation in open models to complement the likes of Claude and Codex, not replace them. Especially with open models, the benchmarks at release are an extremely incomplete story. In some ways this is exciting, as new open models have a much higher variance and ability to surprise, but it also points at some structural reasons that make building businesses and great AI experiences around open models harder than the closed alternatives. When a new Claude Opus or GPT drops, spending a few hours with them in my agentic workflows is genuinely a good vibe test. For open models, putting them through this test is a category error. Something else to be said about open models in the era of agents is that they get out of the debate of integration, harnesses, and tools and let us see close to the ground on what exactly is the ability of just a model. Of course, we can’t test some things like search abilities without some tool, but being able to measure exactly the pace of progress of the model alone is a welcome simplification to a systematically opaque AI space. Share The list of factors I’d use to assess a new open-weight model I’m considering investing in includes: Model performance (and size) — how this model performs on benchmarks I care about and how it compares to other models of a similar size. Country of origin — some businesses care deeply about provenance, and if a model was built in China or not. Model license — if a model needs legal approval for use, uptake will be slower at mid-sized and large companies. Tooling at release — many models release with half-broken, or at least substantially slower, implementations in popular software like vLLM, Transformers, SGLANG, etc due to pushing the envelope of architectures or tools. Model fine-tunability — how easy or hard it is to modify the given model to your use-case when you actually try and use it. The core problem is that some of these are immediately available at release, e.g. general performance, license, origin, etc. but others such as tooling take day(s) to week(s) to stabilize, and others are open research questions — with no group systematically monitoring fine-tunability. In the early era of open models, the days of Llama 2 or 3 and Qwen pre v3.5, the architectures were fairly simple and the models tended to work out of the box. Some of this was due to the extremely hard work of the Llama, Qwen, Mistral, etc. developer teams. Some is due to the new models being genuinely harder to work with. When it comes to something like Qwen 3.5 or Nemotron 3, with hybrid models (either gated delta net or mamba layers), the tooling is very rough at release. Things you would expect to “just work” often don’t. I’ve been following this area closely since we released Olmo Hybrid with a similar architecture, and Qwen 3.5 is just starting to work well in the various open-source tools that need to all play nice together for RL research. That’s 1.5 months after the release date! This is just to start really investing more into understanding the behavior of the models. Of course, others started working on these models sooner by investing more engineering resources or relying on partially closed software. The fully open and distributed ecosystem takes a long time to get going on some new models. All of this is lead-in for the most important question for open models — how easy is it to adapt to specific use-cases? This is a different problem for different model sizes. Large MoE open-weight models may be used by entities like Cursor who need complex capabilities in their domain, e.g. Composer 2 trained on Kimi K2.5. Other applications can be built on much smaller models, such as Chroma’s Context-1 model for agentic search, built on GPT-OSS 20B. The question of “which models are fine-tunable” is largely background knowledge known by engineers across the industry. There should be a thriving research area here to support the open ecosystem model. The first step is to understand characteristics of different base and post-trained models to understand what they look like. The second step is to tune pretraining recipes for open models so they’re more flexible. Interconnects AI is a reader-supported publication. Consider becoming a subscriber. Subscribe For The ATOM Project and other Interconnects endeavors, we’ve put in substantial effort to measuring adoption trends in the open ecosystem. Everything takes a long time to unfold after a model is first publicly available — and adaptability is why. What we know for sure now, when Qwen has been going from strength to strength with its releases, is that technical staff across the industry has gotten comfortable working with Qwen models. Countless research methods and datasets were made to work with Qwen. It’ll take patience for any other model family to get to this point — a patience I’m not sure many open model builders have. This takes us to Gemma 4, Google’s latest open models. Gemma 3 was released more than a year ago, in March of 2025, and is a bit underrated. Gemma 4 comes in 4 sizes for now, with a bigger, MoE model of over 100B total parameters rumored but not released yet. The models we have today come in sizes of ~5B dense, 8B dense, 26B total 4B active MoE, and 31B dense. I’m most excited that they’re finally adopting a standard Apache 2.0 open source license. This’ll massively boost adoption. The standard of better licenses for strong open-weight LLMs was set by mostly Chinese open model labs in the last 1-2 years, and now U.S. companies are following suit. I will personally be so happy if the horrible Llama licenses and Gemma terms of service were an ~18-month transient dynamic of the industry being nervous about releasing strong open models. The Gemma 4 scores look very solid, the small models have incredible benchmark scores (especially in general domains like LMArena) and the 31B model rivals the recent Qwen 3.5 27B, which is the leading member of that class. The ~30B size range is an important one, as it’s accessible both to researchers and to enterprises looking to deploy the model in real use-cases. Where the 7B model scale is the default for tinkering and research, a 30B model is the default for seeing if an open model can unlock substantial value in your specific workflow — a good mix of intelligence, low price, tractability for downstream training, etc. This takes us back to the above adoption criteria I mentioned for open models and the bigger question — do I think Gemma 4 will be an overwhelming success? Previous Gemma models have been plagued by tooling issues and poorer performance when being finetuned. Gemma 4’s success is going to be entirely determined by ease of use, to a point where a 5-10% swing on benchmarks wouldn’t matter at all. It’s strong enough, small enough, with the right license, and from the U.S., so many companies are going to slot it in. I’m cautiously optimistic that Gemma 4 is going to work better here. Winds are shifting for open models built in America. We saw GPT-OSS go through a bumpy launch to become an overwhelming success. There’s a collective energy around the likes of Reflection, Arcee, Nemotron, Gemma, Olmo, and peers that show substantial demand for building new stacks around open models. There’s capital to be spent on AI stacks across the economy by those who want more ownership of everything, including the model. After launching The ATOM Project 240 days ago, the conversation is shifting into the next stage. Summer of 2025 was a crisis moment where the U.S. AI scene realized it can’t wait and figure out open models after building AGI. The two markets will capture different areas and proceed in parallel. Now that more companies in the U.S. are releasing strong models, we need to improve the ecosystem so that these models are easy to use, understand, and build value around. It’s the hard work to build another inflection point in these adoption plots I’ve been updating consistently, but that’s the work to be done. Join me in it. More data coming soon! Here’s a sneak peek:
Editorial: Kaufman Astoria Studios foreclosure fight a warning sign for city’s entertainment industry
The foreclosure battle could spell more trouble for New York's entertainment industry going forward.
If You Identify As A Silly Goose, You Are Going To Love These 26 Products
It's a badge you wear with pride and must honor with this frog soap that wiggles its booty.
Sol and Ky's new specials for Guilty Gear Strive 2.00 detailed, and you're not going to like getting hit by it in the corner or jump
Update: Arc System Works has now revealed more details on Ky Kiske's new special move for Guilty Gear Strive Version 2.00, which you can find added to our story.Guilty Gear Strive is about to receive its Version ...
Last Call for 4.2.26 – A prime-time read of what’s going down in Florida
A digest of the day's politics and policy while the bartender refreshes your drink.
Greektown businesses prepare for Detroit's Opening Day crowds despite ongoing construction
Detroit's Greektown businesses are gearing up for the Opening Day rush on Friday, even though the area remains a construction zone.
Trump speech on Iran war, recent remarks on oil, NATO, daycare costs land with a thud
Trump's meandering speech did little to assuage U.S. allies and world markets concerned about the conflict's ongoing disruptions to the global oil supply.
U.S. small business jobs hold steady in March, Paychex finds
Paychex reports US small business jobs held steady in march with cautious hiring and 2.66% year-over-year wage growth amid ongoing labor challenges.
Bale Dalton considers Artemis program NASA's premier mission for the 2020s and beyond
Former NASA Chief of Staff: 'We're not going just to say that we've been there.'
Pixlel 11 Pro XL renders show Google going all-in with new camera bar
Pixel 11 Pro XL renders complete our preview of this year's non-foldable models.
Spring clothes + accessories that are going out of style, from stylists
From jean shorts to flashy purses, stylists shared which clothes and accessories people can remove from their closets this spring season.
Morning Sports: April 2, 2026
Sheridan/Johnson County HS Spring Sports: Track and Field teams from Sheridan, Big Horn, Tongue River and Buffalo are going to try to beat the weather and compete in a quad meet today at Sheridan. Competition starts at 3pm. In Soccer, Sheridan will play a boys and girls double-header at Worland, while the Buffalo boys and […]
Columbus is betting big on women's sports — and the payoff is already here
Columbus is going all in on women's sports — and the city's bottom line is already feeling it.
'We need to keep it going': Downtown Platte City sees rise in visitors as new businesses open their doors
Downtown Platte City is experiencing new momentum as a newly completed master plan aims to transform the area from a "drive-through" community into a true destination.
Pokemon Home 4.0.0 update released, support for Pokemon Legends: Z-A and Pokemon Champions added
As previously announced, Pokemon Home has received a version 4.0.0 update. Users should be seeing the update roll out now on mobile and Nintendo Switch devices. Keep in mind that the app’s servers are currently undergoing maintenance, making the software unusable until 5 AM PT / 8 AM ET / 1 PM in the UK / 2 PM in Europe....
Oil prices rise as Trump offers no clarity on Iran conflict
President Trump's address on Iran left oil markets unsettled, leading to a spike in Brent and WTI futures amid ongoing geopolitical tensions.
What we know about Wisco's Zavier Zens reopening his recruitment
The 2026 Mr. Basketball winner reopened his recruitment March 30. Here is what to know about his ongoing recruitment.
Last Call for 4.1.26 – A prime-time read of what’s going down in Florida
A digest of the day's politics and policy while the bartender refreshes your drink.
SAVE plan: New email going to millions of student-loan borrowers
The SAVE student-loan repayment is dead, and borrowers will have to find a new plan by the fall. Trump is advising them to switch as soon as possible.
Starcrafty strategy game Stormgate is about to lose its multiplayer modes, due to an AI company buying up Frost Giant's server provider
Stormgate's multiplayer modes are going offline soon as a result of the game's server provider being bought by an AI company, Frost Giant have announced.
The career ladder is fading as AI reshapes work: Linkedin exec
"Workers are going to be climbing a wall, not a ladder," Aneesh Raman, LinkedIn's chief economic opportunity officer, told Business Insider.
Going to Bat: Small business initiative returns for 2026
The Official Site of Minor League Baseball web site includes features, news, rosters, statistics, schedules, teams, live game radio broadcasts, and video clips.
Turn the Page with Paige: ‘St. Louis Sports for Kids’ by Joey Zanaboni
You hear it all the time - St. Louis is a sports town. But it’s only going to stay that way if the next generation learns the love of the game.
Golden Entertainment shareholders approve going-private deal
Golden Entertainment shareholders approve going-private deal
We know what day it is but these Raspberry Pi price hikes are no joke
: Who is going to drop nearly $400 on an underpowered Linux computer?
Cornwall health initiative 'reduces pressure and saves money'
A charity says the project is "transformational" for people going through difficult times.
Where do Newcastle need to strengthen this summer?
Take a long-serving player like Fabian Schar. Newcastle are going to have to be smart with their trading to generate the sorts of funds to enable the club to strengthen the spine of the team, which has to be the priority. Ed and Lara both asked: How likely is it that Sandro Tonali will leave?