SCROLL NEWS / DISCOVERY
Search headlines
Find the stories shaping the conversation.
Results for " k2 " 981 found 🔀 AI-powered Shuffle
These Fun Pursuits Help Middle-Aged Folks Protect Brain Health, Study Finds
By Dennis Thompson HealthDay ReporterFRIDAY, April 24, 2026 (HealthDay News) — Picking up an instrument. Taking a trip abroad. Going out to dinner with friends. All of these are pleasant activities that add spice to life. And they are also some of the most powerful ways to boost brain power and ...
I Mostly Ignored Politics for 10 Months. What Did I Miss?
Trump hasn’t changed. But we have.
SBA Offers Relief to Oklahoma Small Businesses
U.S. Small Business Administration announces low interest disaster loans
Suns-Thunder takeaways: There doesn't appear to be an answer for SGA
Youssef Zalal aims to end Aljamain Sterling's 'point-proving streak'
Just days ahead of his first UFC main event, Youssef Zalal can feel he's on the verge of big things in his career.
Kroenke Sports & Entertainment to renovate 24 suites at Ball Arena
Kroenke Sports & Entertainment is planning to renovate 24 luxury suites at Ball Arena during the summer and debuting in time for the 2026-27 Nuggets and Avalanche seasons.
Stumbling upon science fiction with ‘I, Robot’
When I was growing up, my father had a bookshelf with glass doors. Behind the delicate handles were elegant hardcovers, fairy tale collections with beautiful illustrations and sentimental classics, li...
UGREEN Enters Sports Arena with New England Patriots Partnership
UGREEN, a global leader in consumer technology, today announced a partnership with the New England Patriots, becoming the Official Technology Accessory Partner for the team’s 2026 season.
Taking Vitamin D? Magnesium and K2 Could Make a Difference
Should you take vitamin D with magnesium and vitamin K2? Here’s how these supplements work together, plus safety tips and recommended doses.
Bullying, Politics Harm Mental Health Of Gender-Diverse Teens
By Dennis Thompson HealthDay ReporterWEDNESDAY, April 22, 2026 (HealthDay News) — Gender-diverse teenagers who are bullied are more likely to suffer escalating psychological distress than other teens, particularly if they live in a state with repressive gender identity laws, a new study says. These ...
Voters Approving Virginia Democratic Gerrymander Puts Democrats on Cusp of 218 House Seats in Ratings, but Questions Remain
Four House ratings move in Democrats' favor
Euclid Space Warps – help us hunt for galaxy-galaxy lense
“Space Warps,” the new citizen science project on the Zooniverse platform, aims to identify gravitational lensing in Euclid data and study dark matter.
Salmon contract signing another great step for young defender
Arsenal have announced that 16-year-old centre back Marli Salmon has signed a pre-contract agreement which will mean he signs a first professional contract when he turns 17 in August. Salmon was taken on the tour of Hong Kong and Singapore and impressed the coaching staff, featuring in games against Newcastle and AC Milan. The centre-back made his debut in the Champions League in the 3-0 win over Club Brugge and has also made three appearances in the FA Cup, starting the win over Mansfield.
Microsoft Teams is trying to fix accidental hand-raising
The planned redesign will also allow users to customize their actions toolbar.
Victor Wembanyama named NBA Defensive Player of the Year award as first ever unanimous winner
Business Monday: Recovery Center to help small businesses affected by downtown Hilo fire
The U.S. Small Business Administration is setting up a recovery center where representatives will help businesses that burned in the 2025 Hilo fire apply for low-interest Economic Injury Disaster Loans.
iPhone 18 Pro Launching in September With These 10 New Features
While the iPhone 18 Pro and iPhone 18 Pro Max are not launching until September, there are already plenty of rumors about the devices. It was initially reported that the iPhone 18 Pro models would have fully under-screen Face ID, with only a front camera visible in the top-left corner of the screen. However, the latest rumors indicate that only one Face ID component will be moved under the screen on the devices, which will result in merely a smaller Dynamic Island.
‘We aren’t going anywhere’: Bama Health Foods to stay open after vehicle crashes into building
Officials say despite some of the store being boarded up and a portion of it missing, they will remain open.
2 people shot, 1 seriously injured at Birmingham business Sunday morning
Police said that no one is in custody at this time.
Defender knocks back Barcelona contract offer
Barcelona are gearing up for a huge summer, but before the transfer window opens, they have a number of contract situations to resolve. The big one centres on Robert Lewandowski, but another whose cur...
Entertainment figures reject calls for Israel Eurovision boycott
Public letter says kicking the Jewish state out of song contest would be "inversion of justice."
Business: Rewarding Year-rounders, Chatham Business Offers Incentives For Residents, Businesses
Even though the shoulder seasons have become busier in recent years, Lower Cape businesses can still struggle to remain open year-round. Lindsay...
Jose Mourinho Has Entered the Newcastle Conversation
Newcastle United are approaching decision time on Eddie Howe’s future as the manager of the side. An awful end to the season and a team that cannot escape their never-ending struggles, currently sitting in the very lower half of the Premier League table, simply screams for action. Howe remains in charge and, seemingly, for some […]
What is ‘Ozempic personality,’ and why does it make life feel ‘meh’?
Doctors and patients across social media are reporting a kind of emotional flattening while on GLP-1 drugs.
Dodgers vs. Mets game III chat
The Dodgers conclude their three-game series against the New York Mets on Jackie Robinson Day, looking to get a sweep and take five of six games on the homestand. Shohei Ohtani makes his third start of the season against right-hander Clay Holmes. Ohtani will only be on the mound on Wednesday, as Dalton Rushing is […]
Greater Wausau Chamber set to host Business EXPO Thursday
The Greater Wausau Chamber of Commerce will host hundreds of exhibitors and thousands of attendees at the organization’s annual Business EXPO on Thursday, April 16.
Kent State football coach Mark Carney receives Josh Cribbs' approval
Former Cleveland Browns star Josh Cribbs visited Kent State University and shared thoughts on the state of his college team under coach Mark Carney.
Simon Cherry Receives ESMI Award for Significant, Long-Term Contributions to Imaging Science
The Distinguished Research Professor of biomedical engineering has received the 2026 ESMI Award from the European Society for Molecular Imaging. Among the most prestigious honors in imaging science, the award celebrates a researcher for their outstanding contributions to the interdisciplinary field.
Tim Wilkerson to be inducted into Illinois Sports Hall of Fame Class of 2026
The Illinois Sports Hall of Fame will induct NHRA Funny Car racer Tim Wilkerson into its Class of 2026, honoring his outstanding career and lasting impact on the sport of drag racing.
Minneapolis public safety debate back in a familiar place: anger, political squabbling
The Minneapolis City Council’s rejection of Mayor Jacob Frey’s public safety commissioner renewed the angry divisions over policing that have characterized city politics in recent years.
The Dirt: Frontier Behavioral Health planning new Spokane location
The longstanding location of Spokane Roofing Company will soon become a Frontier Behavioral Health location, according to a construction permit application submitted to the city of Spokane.
“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.
Silver State Sports LIVE
Catch the Silver State Sports & Entertainment Network (SSSEN) over the air on 5.2, COX 77, YouTubeTV, and Fubo.
Trump small business admin Kelly Loeffler talks tariffs, taxes, gas prices
Small Business Administrator Kelly Loeffler was back in Georgia to tour a factory Thursday afternoon.
LeBron James dunks right behind KTVU sports reporter Bailey O'Carroll
It's all eyes on Lakers LeBron James ahead of the Warriors final home game of the regular season. While warming up with some free throws ahead of their game with the Warriors, LeBron James decides to dunk, nearly taking out our sports reporter!
Masters 2026 leaderboard, tee times, how to watch: 60-year-old Jose Maria Olazabal turning back the clock
Jolene Jolene sports bar finds permanent Atlanta home
Atlanta women’s sports bar Jolene Jolene secured a permanent home after two years of local pop-ups.
Masters leaderboard, tee times, how to watch: Keep up with Rory McIlroy, Scottie Scheffler at Augusta National
Gemini gets notebooks to help you organize projects
Notebooks sync back with NotebookLM, too.
Huck Distinguished Lecture Series brings leaders in AI, biodiversity and cryo‑EM
The Huck Distinguished Lecture Series in April will feature Tanya Berger-Wolf of The Ohio State University and Wah Chiu of Stanford University, highlighting advances in artificial intelligence for biodiversity research and cryogenic electron microscopy for biomedical discovery.
Baton Rouge General's Health Tip of the Week: What to know about new cholesterol guidelines
Sponsored by Baton Rouge General Heart health is getting a head start—and the new guidelines say your 30s are the time to begin. Updated recommendations call for annual cholesterol screenings starting at age 30, lower LDL (“bad” cholesterol) targets (under 100, or 70 if you’re higher risk), and additional tests like lipoprotein(a) at least once. […]
MedTech Entrepreneurs Secure a Leg Up with PLASMA Health at Aggie Square
There’s a new educational arm to PLASMA, the flagship entrepreneurship program of the Student Startup Center. The 12-week accelerator, which helps undergraduates launch a startup business, now features a track for students interested in medtech.
Brenda Alexander has the latest in entertainment headlines:
Are Kim K & Lewis Hamilton dating? Lil Nas X may have caught a break in his felony assault case, and Stephen Colbert's replacement has been announced.
State’s first Health Learning Lab opens at Waianae High School
The lab provides students with hands-on clinical skills training in a modern setting and supports certificate programs designed to lead directly to entry-level healthcare jobs.
Melissa Gilbert says child abuse allegations ruined her life with Timothy Busfield: 'He's canceled'
Melissa Gilbert defended husband Timothy Busfield against child abuse allegations in her first TV interview on 'Good Morning America,' saying the 'traumatic' experience ruined their future: 'It's done. He's canceled.'
DitD & Open Post - 4/6/26: First Pitch Edition
New Jersey Devils & Related Hockey Links for 4/6/26
Geno Auriemma's latest meltdown involving South Carolina's Dawn Staley comes with a warning sign for his legacy
Attacks against hospitals and health facilities must end
WHO, UNICEF raise alarm over escalating attacks on health care in Sudan
Cluster Munitions and the Politics of Enforcement
The Use of Cluster Munitions Doesn’t Draw the Same Scrutiny When Israel or Israeli Civilians Are the Victims of Attacks
President Trump signs executive order intended to stabilize college sports
In the order signed hours before the women's Final Four tipped off one of the biggest weekends in college sports, President Donald Trump went after eligibility rules, transfers and the spiraling costs associated with an industry that now pays its players millions of dollars per year.
The Happy Accident: How Benetta MacAuley Found Her Joy in Materials Science and Music
Benetta MacAuley didn't know materials science was a major when she applied to UC Davis. Now the senior investigates sustainable cement alternatives and leads the viola section of the Video Game Orchestra, proving that the best paths are often unplanned.
President Trump expected to deliver new executive order attempting to regulate college sports: Sources
Sources: President Trump expected to deliver new executive order attempting to regulate college sports
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:
Materials Science and Engineering Weekly Seminar with Dr. Mathias Kläui
Combining The Best Of All Worlds: Altermagnets – A New Class Of MagnetsAbstract While known for a long time, antiferromagnetically ordered systems have previously been considered as “interesting but useless”.
GTA+ Members Can Claim a Free Bravado Buffalo STX Pursuit Law Enforcement Vehicle
Get behind the wheel of the aggressive Bravado Buffalo STX Pursuit Law Enforcement Vehicle now available to GTA+ Members one week before everyone else on April 9. Enjoy other great benefits including bonuses for Dispatch Work and Bail Office Bounties,...
Giants-Mets Series Preview: Good vs. Good enough
The Mets radically reinvented themselves in the offseason while the Giants tinkered. Which strategy will win?
Cisco Patches 9.8 CVSS IMC and SSM Flaws Allowing Remote System Compromise
Cisco patches two 9.8 CVSS flaws (CVE-2026-20093, CVE-2026-20160), preventing authentication bypass and root access.
Exclusive: Google Pixel 11 Pro XL Official CAD Renders & Rumors
After showing off the Pixel 11 and Pixel 11 Pro earlier this week, let's end the week on a high note with the Pixel 11 Pro XL. As expected, the Pixel 11
USC Big Ten ranking in one preseason prediction model unveiled
The new offseason SP+ rankings are out. See where USC landed in this first preseason measurement of the Big Ten's 18 programs:
UC Davis Engineers Help Advance Collaboration, Innovation with New AI and Health Award
The AI and Health Seed Funding Program is a new initiative focused on strengthening interdisciplinary connections between schools and colleges across UC Davis and UC Davis Health. Engineers are members of three out of six teams to receive the program’s inaugural awards.
Exclusive: Google Pixel 11 Pro Official CAD Renders & Rumors
The Google Pixel 11 Pro is set to debut later this summer, alongside the Google Pixel 11 which we leaked yesterday. The design is staying basically the
Apple to Launch These 15+ New Products Later This Year
March has been an incredibly busy month for Apple, with the company unveiling more than 10 new products and accessories. We said hello to the MacBook Neo at the start of the month, and we bid farewell to the Mac Pro at the end of it. Beyond the usual annual updates to iPhones and Apple Watches, Apple's all-new smart home hub is finally expected to launch later this year, once the more personalized version of Siri arrives.
AitM Phishing Targets TikTok Business Accounts Using Cloudflare Turnstile Evasion
AitM phishing hijacks TikTok Business accounts via Cloudflare Turnstile evasion as SVG malware linked to BianLian targets Venezuela.
Italian regulator launches probe into LVMH, Sephora for unfair commercial practices
The Italian competition authority AGCM said on Friday it had launched two investigations into LVMH-owned entities in Italy, citing unfair commercial practices linked to the premature use of adult cosmetics among minors.
Breaking down barriers to mental health support
By removing stigmatizing credentialing questions and creating a peer support network, Bayhealth is aiming to help physicians feel comfortable seeking the behavioral health care they need.
Google is making it easier to import another AI’s memory into Gemini
Prompt your AI to copy-paste its memory into Gemini.
Iran bans sports teams from travelling to countries it deems 'hostile'
Iran has banned national and club sports teams from travelling to countries it considers hostile until further notice, Iranian media reported on Thursday citing the Sports Ministry, which said the move was due to concerns over the safety of Iranian athletes.
House Democrat accused of misspending covid-19 money could be expelled
The family health care company Rep. Sheila Cherfilus-McCormick worked for was sent FEMA money it shouldn’t have received. She’s accused of using the money on a run for office.
Exclusive: Samsung Galaxy Z Fold 8 Wide Official CAD Renders & Rumors
Samsung is planning a second book-style fold for later this year, specifically to compete with the upcoming iPhone Fold. Apple is said to be launching a
Scientists discover mirror of our solar system in 2 exoplanets forming around a star
Astronomers have spotted two planets forming around a distant infant star, offering a time capsule into see what the solar system looked like billions of years ago.
'Sports Illustrated Swimsuit' Teases Sexy New Photo Shoots with Bethenny Frankel, Ilona Maher and More
'Sports Illustrated Swimsuit' teased its forthcoming 2026 issue featuring a mix of newcomers, as well as fan-favorite veterans.
Why REI treats CX as a critical business function
The outdoor retailer is focused on doubling down on its differentiators, REI’s Clay Walton-House said. That means investing in its elevated services and member experience.
Bachelorette suitor from canceled season drove car in Madison Cawthorn crash
A 'Bachelorette' suitor from Taylor Frankie Paul's now-scrapped season was the driver in the 2014 car crash that paralyzed former Republican congressman Madison Cawthorn.
Rock and roll icon to perform in central Pa. concert this summer
The tour is in support of his new album, “Automatic,” his first original release in five years.
Cursor acknowledges its new low-cost coding model has Chinese roots
Cursor says its new coding model builds on Kimi K2.5, a Chinese model it didn't mention at launch.
Lakers player grades: Luke Kennard gives L.A. ninth straight win
The Lakers defeated the Magic for their ninth straight win thanks to Luke Kennard's heroics at the end of the game.
Allina-Sutter deal will likely boost AI in Minnesota health care
Contractor who tests accuracy of AI tools used by both health systems said they are liked-minded in their ambitions to improve care through technology.
Argyle record-breaker Campbell signs first pro deal
Plymouth Argyle's youngest-ever goalscorer Seb Campbell signs his first professional deal with the club.
Hometown Health: Virginia teacher continues career through virtual academy after MS diagnosis
Suzy Mullins spent years not knowing what was happening to her body
Gavin Adcock ‘The Day I Hang It Up Tour’ includes two Alabama shows: Act fast and secure your tickets
Fans can catch Adcock at popular venues in Orange Beach and Birmingham later this year.
I had to jump in and celebrate with fans
Ryley Towler thought only of celebrating with Lincoln City fans in the stands when he scored the equaliser at Huddersfield.
Douglas County Commissioner Mary Ann Borgeson reflects on 32 years in politics
As Women’s History Month highlights the achievements of women in society and culture, one woman has served the metro area community for more than three decades.
Chelsea vs PSG – Match preview and team news
Chelsea welcome Paris Saint-Germain to Stamford Bridge tonight, needing a miracle to overturn a daunting three-goal deficit against the Champions League holders.The Blues arrive following a bruising 5...
Apple Acquires Final Cut Pro Plugin Company MotionVFX
Apple today acquired the major Final Cut Pro plugin company MotionVFX. MotionVFX is a Polish company founded by Szymon Masiak in 2009. It creates high-quality plugins, transitions, templates, and visual effects (VFX) for video editors. The company specializes in tools for Final Cut Pro, DaVinci Resolve, and Apple Motion, such as mFilmLook, mO2, and the Design Studio extension.
A 79-Year-Old Freshman Senator?
Janet Mills is sharp, but her candidacy is haunted by Joe Biden’s 2024 debacle.
Chargers free agency: LA hosting offensive lineman on visit
The Los Angeles Chargers are looking to continue building out the interior part of the offensive line.
From flat jokes, to politics, to red-carpet dazzle: 5 experts unpack the 2026 Oscars
Between the glitzy celebration of cinematic escapism, and constant reminders of our dark and complicated world, this year’s ceremony felt somewhat disjointed.
Lakers’ LeBron James calls Luka Dončić ‘generational player’
Luka Dončić made his first game-winner with Lakers in 127-125 overtime win over the Denver Nuggets on Saturday at Crypto.com Are
Spirit Airlines to shrink fleet to one‑third of pre‑bankruptcy size
Spirit Aviation Holdings , the parent company of Spirit Airlines, said on Friday it plans to shrink its fleet to about one‑third of its pre‑bankruptcy size, according to a court filing.
How have 'belief and clarity' improved Ulster?
After Friday's win in Edinburgh took them to second in the table, how have "belief and clarity" helped Ulster's resurgence?