At Gamescom Latam, Roblox announced it has broken ground on a new data center in Brazil — slated to go live in early 2026.Read More
Roblox breaks ground on data center in Brazil for early 2026 Read Post »
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. A senior State Department official demanded records of communications with journalists, European officials, and Trump critics A previously unreported document distributed by senior US State Department official Darren Beattie reveals a sweeping effort to uncover all communications between the staff of a small government office focused on online disinformation and a lengthy list of public and private figures—many of whom are longtime targets of the political right. The document, originally shared in person with roughly a dozen State Department employees in early March, requested staff emails and other records with or about a host of individuals and organizations that track or write about foreign disinformation—including Atlantic journalist Anne Applebaum, former US cybersecurity official Christopher Krebs, and the Stanford Internet Observatory—or have criticized President Donald Trump and his allies, such as the conservative anti-Trump commentator Bill Kristol. The broad requests for unredacted information felt like a “witch hunt,” one official says—one that could put the privacy and security of numerous individuals and organizations at risk. Read the full story. —Eileen Guo The US has approved CRISPR pigs for food Most pigs in the US are confined to factory farms where they can be afflicted by a nasty respiratory virus that kills piglets. The illness is called porcine reproductive and respiratory syndrome, or PRRS. A few years ago, a British company called Genus set out to design pigs immune to this germ using CRISPR gene editing. Not only did they succeed, but its pigs are now poised to enter the food chain following approval of the animals this week by the U.S. Food and Drug Administration. Read the full story. —Antonio Regalado This article is from The Checkup, MIT Technology Review’s weekly health and biotech newsletter. To receive it in your inbox every Thursday, sign up here. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 The US has closed a China tariff loopholeThe costs of plenty of goods are likely to shoot up in response. (NYT $)+ But China is still extremely dependent on US-made car chips. (WSJ $)+ Chinese retail giant Temu is pivoting its business model. (Bloomberg $)+ Sweeping tariffs could threaten the US manufacturing rebound. (MIT Technology Review) 2 DOGE’s future is looking uncertainIt’s fallen far short of its goal to slash $2 trillion in spending. (WP $)+ No more late-night ice cream for Elon Musk. (CNBC)+ DOGE’s tech takeover threatens the safety and stability of our critical data. (MIT Technology Review) 3 Microsoft is hiking the price of its Xbox games consoleBy a whopping 27% in the US. (The Guardian)+ Apple estimates that the tariffs will add $900 million to its costs. (WP $)+ But Apple isn’t announcing any price increases (yet.) (TechCrunch)+ Here’s what is—and isn’t—getting pricier under the tariffs. (Vox) 4 Tech giants have been accused of deliberately distorting AI rankingsA new study claims they’re making untrue claims about the best models. (New Scientist $)+ It accuses benchmark organisation LM Arena of unfair practices. (TechCrunch)+ The site’s operators refute the findings, saying its conclusions are wrong. (Ars Technica) 5 Europe wants to replicate America’s military-industrial complexAnd US contractors are likely to benefit. (WSJ $)+ US soldiers may finally be able to repair their own equipment. (404 Media)+ Generative AI is learning to spy for the US military. (MIT Technology Review) 6 Elon Musk’s lawsuit against OpenAI will move forwardA judge rejected OpenAI’s attempt to dismiss the case. (FT $) 7 What a post-4Chan internet looks likeWhat was once contained to a tiny corner of the web is now commonplace. (New Yorker $)+ How to fix the internet. (MIT Technology Review) 8 How North Korea infiltrates the USFully remote coders are not who they appear to be. (Wired $) 9 You no longer need a password to open a new Microsoft accountThe company’s gone passkey-first. (The Verge) 10 Fecal transplants are a possible way to treat gut disease And the approach is becoming more mainstream. (Undark)+ How bugs and chemicals in your poo could give away exactly what you’ve eaten. (MIT Technology Review) Quote of the day “What about the next Taylor Swift?” —US District Court Judge Vince Chhabria questions how powerful musical AI tools will affect up-and-coming musicians during Meta’s copyright court battle, Wired reports. One more thing Your boss is watching Working today—whether in an office, a warehouse, or your car—can mean constant electronic surveillance with little transparency, and potentially with livelihood-ending consequences if your productivity flags. But what matters even more than the effects of this ubiquitous monitoring on privacy may be how all that data is shifting the relationships between workers and managers, companies and their workforce. We are in the midst of a shift in work and workplace relationships as significant as the Second Industrial Revolution of the late 19th and early 20th centuries. And new policies and protections may be necessary to correct the balance of power. Read the full story. —Rebecca Ackermann We can still have nice things A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.) + This is cool: scientists have successfully triggered a lightning strike using a drone. + It’s the age-old question—why do so many men refuse to wear shorts in hot weather?+ The American accent that’s hardest for British actors to pull off seems to be either New York or Boston.+ Happy 50th birthday to David Beckham, best of British.
The Download: foreign disinformation intel, and gene-edited pork Read Post »
arXiv:2505.00367v1 Announce Type: new Abstract: Cognitive distortion refers to negative thinking patterns that can lead to mental health issues like depression and anxiety in adolescents. Previous studies using natural language processing (NLP) have focused mainly on small-scale adult datasets, with limited research on adolescents. This study introduces KoACD, the first large-scale dataset of cognitive distortions in Korean adolescents, containing 108,717 instances. We applied a multi-Large Language Model (LLM) negotiation method to refine distortion classification and generate synthetic data using two approaches: cognitive clarification for textual clarity and cognitive balancing for diverse distortion representation. Validation through LLMs and expert evaluations showed that while LLMs classified distortions with explicit markers, they struggled with context-dependent reasoning, where human evaluators demonstrated higher accuracy. KoACD aims to enhance future research on cognitive distortion detection.
KoACD: The First Korean Adolescent Dataset for Cognitive Distortion Analysis Read Post »
arXiv:2504.19467v2 Announce Type: replace Abstract: Large language models (LLMs) hold great promise for medical applications and are evolving rapidly, with new models being released at an accelerated pace. However, current evaluations of LLMs in clinical contexts remain limited. Most existing benchmarks rely on medical exam-style questions or PubMed-derived text, failing to capture the complexity of real-world electronic health record (EHR) data. Others focus narrowly on specific application scenarios, limiting their generalizability across broader clinical use. To address this gap, we present BRIDGE, a comprehensive multilingual benchmark comprising 87 tasks sourced from real-world clinical data sources across nine languages. We systematically evaluated 52 state-of-the-art LLMs (including DeepSeek-R1, GPT-4o, Gemini, and Llama 4) under various inference strategies. With a total of 13,572 experiments, our results reveal substantial performance variation across model sizes, languages, natural language processing tasks, and clinical specialties. Notably, we demonstrate that open-source LLMs can achieve performance comparable to proprietary models, while medically fine-tuned LLMs based on older architectures often underperform versus updated general-purpose models. The BRIDGE and its corresponding leaderboard serve as a foundational resource and a unique reference for the development and evaluation of new LLMs in real-world clinical text understanding. The BRIDGE leaderboard: https://huggingface.co/spaces/YLab-Open/BRIDGE-Medical-Leaderboard
Most pigs in the US are confined to factory farms where they can be afflicted by a nasty respiratory virus that kills piglets. The illness is called porcine reproductive and respiratory syndrome, or PRRS. A few years ago, a British company called Genus set out to design pigs immune to this germ using CRISPR gene editing. Not only did they succeed, but its pigs are now poised to enter the food chain following approval of the animals this week by the U.S. Food and Drug Administration. The pigs will join a very short list of gene-modified animals that you can eat. It’s a short list because such animals are expensive to create, face regulatory barriers, and don’t always pay off. For instance, the US took about 20 years to approve a transgenic salmon with an extra gene that let it grow faster. But by early this year its creator, AquaBounty, had sold off all its fish farms and had only four employees—none of them selling fish. Regulations have eased since then, especially around gene editing, which tinkers with an animal’s own DNA rather than adding to it from another species, as is the case with the salmon and many GMO crops. What’s certain is that the pig project was technically impressive and scientifically clever. Genus edited pig embryos to remove the receptor that the PRRS virus uses to enter cells. No receptor means no infection. According to Matt Culbertson, chief operating office of the Pig Improvement Company, a Genus subsidiary, the pigs appear entirely immune to more than 99% of the known versions of the PRRS virus, although there is one rare subtype that may break through the protection. This project is scientifically similar to the work that led to the infamous CRISPR babies born in China in 2018. In that case a scientist named He Jiankui edited twin girls to be resistant to HIV, also by trying to remove a receptor gene when they were just embryos in a dish. That experiment on humans was widely decried as misguided. But pigs are a different story. The ethical concerns about experimenting are less serious, and the benefits of changing the genomes can be measured in dollars and cents. It’s going to save a lot of money if pigs are immune to the PRRS virus, which spreads quite easily, causing losses of $300 million a year or more in the US alone. Globally, people get animal protein mostly from chickens, with pigs and cattle in second and third place. A 2023 report estimated that pigs account for 34% of all meat that’s eaten. Of the billion pigs in the world, about half are in China; the US comes in a distant second, with 80 million. Recently, there’s been a lot of fairly silly news about genetically modified animals. A company called Colossal Biosciences used gene editing to modify wolves in ways it claimed made them resemble an extinct species, the dire wolf. And then there’s the L.A. Project, an effort run by biohackers who say they’ll make glow-in-the-dark rabbits and have a stretch goal of creating a horse with a horn—that’s right, a unicorn. Both those projects are more about showmanship than usefulness. But they’re demonstrations of the growing power scientists have to modify mammals, thanks principally to new gene-editing tools combined with DNA sequencing that lets them peer into animals’ DNA. Stopping viruses is a much better use of CRISPR. And research is ongoing to make pigs—as well as other livestock—invulnerable to other infections, including African swine fever and influenza. While PRRS doesn’t infect humans, pig and bird flus can. But if herds and flocks could be changed to resist those infections, that could cut the chances of the type of spillover that can occasionally cause dangerous pandemics. There’s a chance the Genus pigs could turn out to be the most financially valuable genetically modified animal ever created—the first CRISPR hit product to reach the food system. After the approval, the company’s stock value jumped up by a couple of hundred million dollars on the London Stock Exchange. But there is still a way to go before gene-edited bacon appears on shelves in the US. Before it makes its sales pitch to pig farms, Genus says, it needs to also gain approval in Mexico, Canada, Japan and China which are big export markets for American pork. Culbertson says gene-edited pork could appear in the US market sometime next year. He says the company does not think pork chops or other meat will need to carry any label identifying it as bioengineered. “We aren’t aware of any labelling requirement,” Culbertson says. This article is from The Checkup, MIT Technology Review’s weekly health and biotech newsletter. To receive it in your inbox every Thursday, sign up here.
The US has approved CRISPR pigs for food Read Post »
arXiv:2501.18265v2 Announce Type: replace-cross Abstract: Evaluating the truthfulness of online content is critical for combating misinformation. This study examines the efficiency and effectiveness of crowdsourced truthfulness assessments through a comparative analysis of two approaches: one involving full-length webpages as evidence for each claim, and another using summaries for each evidence document generated with a large language model. Using an A/B testing setting, we engage a diverse pool of participants tasked with evaluating the truthfulness of statements under these conditions. Our analysis explores both the quality of assessments and the behavioral patterns of participants. The results reveal that relying on summarized evidence offers comparable accuracy and error metrics to the Standard modality while significantly improving efficiency. Workers in the Summary setting complete a significantly higher number of assessments, reducing task duration and costs. Additionally, the Summary modality maximizes internal agreement and maintains consistent reliance on and perceived usefulness of evidence, demonstrating its potential to streamline large-scale truthfulness evaluations.
arXiv:2505.00001v1 Announce Type: new Abstract: Large Language Models (LLMs) are primarily trained on high-resource natural languages, limiting their effectiveness in low-resource settings and in tasks requiring deep logical reasoning. This research introduces Rosetta-PL, a benchmark designed to evaluate LLMs’ logical reasoning and generalization capabilities in a controlled environment. We construct Rosetta-PL by translating a dataset of logical propositions from Lean into a custom logical language, which is then used to fine-tune an LLM (e.g., GPT-4o). Our experiments analyze the impact of the size of the dataset and the translation methodology on the performance of the model. Our results indicate that preserving logical relationships in the translation process significantly boosts precision, with accuracy plateauing beyond roughly 20,000 training samples. These insights provide valuable guidelines for optimizing LLM training in formal reasoning tasks and improving performance in various low-resource language applications.
Rosetta-PL: Propositional Logic as a Benchmark for Large Language Model Reasoning Read Post »
China has once again beat everyone else to a clean energy milestone—its new nuclear reactor is reportedly one of the first to use thorium instead of uranium as a fuel and the first of its kind that can be refueled while it’s running. It’s an interesting (if decidedly experimental) development out of a country that’s edging toward becoming the world leader in nuclear energy. China has now surpassed France in terms of generation, though not capacity; it still lags behind the US in both categories. But one recurring theme in media coverage about the reactor struck me, because it’s so familiar: This technology was invented decades ago, and then abandoned. You can basically copy and paste that line into countless stories about today’s advanced reactor technology. Molten-salt cooling systems? Invented in the mid-20th century but never commercialized. Same for several alternative fuels, like TRISO. And, of course, there’s thorium. This one research reactor in China running with an alternative fuel says a lot about this moment for nuclear energy technology: Many groups are looking into the past for technologies, with a new appetite for building them. First, it’s important to note that China is the hot spot for nuclear energy right now. While the US still has the most operational reactors in the world, China is catching up quickly. The country is building reactors at a remarkable clip and currently has more reactors under construction than any other country by far. Just this week, China approved 10 new reactors, totaling over $27 billion in investment. China is also leading the way for some advanced reactor technologies (that category includes basically anything that deviates from the standard blueprint of what’s on the grid today: large reactors that use enriched uranium for fuel and high-pressure water to keep the reactor cool). High-temperature reactors that use gas as a coolant are one major area of focus for China—a few reactors that use this technology have recently started up, and more are in the planning stages or under construction. Now, Chinese state media is reporting that scientists in the country reached a milestone with a thorium-based reactor. The reactor came online in June 2024, but researchers say it recently went through refueling without shutting down. (Conventional reactors generally need to be stopped to replenish the fuel supply.) The project’s lead scientists shared the results during a closed meeting at the Chinese Academy of Sciences. I’ll emphasize here that this isn’t some massive power plant: This reactor is tiny. It generates just two megawatts of heat—less than the research reactor on MIT’s campus, which rings in at six megawatts. (To be fair, MIT’s is one of the largest university research reactors in the US, but still … it’s small.) Regardless, progress is progress for thorium reactors, as the world has been entirely focused on uranium for the last 50 years or so. Much of the original research on thorium came out of the US, which pumped resources into all sorts of different reactor technologies in the 1950s and ’60s. A reactor at Oak Ridge National Laboratory in Tennessee that ran in the 1960s used Uranium-233 fuel (which can be generated when thorium is bombarded with radiation). Eventually, though, the world more or less settled on a blueprint for nuclear reactors, focusing on those that use Uranium-238 as fuel and are cooled by water at a high pressure. One reason for the focus on uranium for energy tech? The research could also be applied to nuclear weapons. But now there’s a renewed interest in alternative nuclear technologies, and the thorium-fueled reactor is just one of several examples. A prominent one we’ve covered before: Kairos Power is building reactors that use molten salt as a coolant for small nuclear reactors, also a technology invented and developed in the 1950s and ’60s before being abandoned. Another old-but-new concept is using high-temperature gas to cool reactors, as X-energy is aiming to do in its proposed power station at a chemical plant in Texas. (That reactor will be able to be refueled while it’s running, like the new thorium reactor.) Some problems from decades ago that contributed to technologies being abandoned will still need to be dealt with today. In the case of molten-salt reactors, for example, it can be tricky to find materials that can withstand the corrosive properties of super-hot salt. For thorium reactors, the process of transforming thorium into U-233 fuel has historically been one of the hurdles. But as early progress shows, the archives could provide fodder for new commercial reactors, and revisiting these old ideas could give the nuclear industry a much-needed boost. This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here.
A long-abandoned US nuclear technology is making a comeback in China Read Post »
arXiv:2504.21026v1 Announce Type: new Abstract: With the growing presence of multilingual users on social media, detecting abusive language in code-mixed text has become increasingly challenging. Code-mixed communication, where users seamlessly switch between English and their native languages, poses difficulties for traditional abuse detection models, as offensive content may be context-dependent or obscured by linguistic blending. While abusive language detection has been extensively explored for high-resource languages like English and Hindi, low-resource languages such as Telugu and Nepali remain underrepresented, leaving gaps in effective moderation. In this study, we introduce a novel, manually annotated dataset of 2 thousand Telugu-English and 5 Nepali-English code-mixed comments, categorized as abusive and non-abusive, collected from various social media platforms. The dataset undergoes rigorous preprocessing before being evaluated across multiple Machine Learning (ML), Deep Learning (DL), and Large Language Models (LLMs). We experimented with models including Logistic Regression, Random Forest, Support Vector Machines (SVM), Neural Networks (NN), LSTM, CNN, and LLMs, optimizing their performance through hyperparameter tuning, and evaluate it using 10-fold cross-validation and statistical significance testing (t-test). Our findings provide key insights into the challenges of detecting abusive language in code-mixed settings and offer a comparative analysis of computational approaches. This study contributes to advancing NLP for low-resource languages by establishing benchmarks for abusive language detection in Telugu-English and Nepali-English code-mixed text. The dataset and insights can aid in the development of more robust moderation strategies for multilingual social media environments.
arXiv:2504.21012v1 Announce Type: new Abstract: What underlies intuitive human thinking? One approach to this question is to compare the cognitive dynamics of humans and large language models (LLMs). However, such a comparison requires a method to quantitatively analyze AI cognitive behavior under controlled conditions. While anecdotal observations suggest that certain prompts can dramatically change LLM behavior, these observations have remained largely qualitative. Here, we propose a two-part framework to investigate this phenomenon: a Transition-Inducing Prompt (TIP) that triggers a rapid shift in LLM responsiveness, and a Transition Quantifying Prompt (TQP) that evaluates this change using a separate LLM. Through controlled experiments, we examined how LLMs react to prompts embedding two semantically distant concepts (e.g., mathematical aperiodicity and traditional crafts)–either fused together or presented separately–by changing their linguistic quality and affective tone. Whereas humans tend to experience heightened engagement when such concepts are meaningfully blended producing a novel concept–a form of conceptual fusion–current LLMs showed no significant difference in responsiveness between semantically fused and non-fused prompts. This suggests that LLMs may not yet replicate the conceptual integration processes seen in human intuition. Our method enables fine-grained, reproducible measurement of cognitive responsiveness, and may help illuminate key differences in how intuition and conceptual leaps emerge in artificial versus human minds.
