Sam Altman, leader of one of the world's biggest artificial intelligence companies, has signed a deal with the British government to explore the deployment of advanced AI models in areas including justice, security and education. The chief executive of OpenAI, which has been valued at 300bn ( 220bn) and provides the ChatGPT suite of large language models, agreed the memorandum of understanding with the science and technology secretary, Peter Kyle, on Monday. It follows a similarly wide-ranging deal between the UK government and OpenAI's rival US tech company, Google, which campaigners called "dangerously naive", citing fears that the arrangement could leave the public sector dependent on private technology providers and make it harder for politicians to regulate them. The latest agreement states that OpenAI and the government "will collaborate to identify opportunities for how advanced AI models can be deployed throughout government", including "to help civil servants work more efficiently" and to support "citizens to navigate public services more effectively". It said they will collaborate to develop AI solutions "to the UK's hardest problems, including in areas such as justice, defence and security, and education technology" and develop partnerships "to expand public engagement with AI technology".

"Technology happens because it is possible," OpenAI CEO, Sam Altman, told the New York Times in 2019, consciously paraphrasing Robert Oppenheimer, the father of the atomic bomb. Another widespread techie conviction is that the first human-level AI – also known as artificial general intelligence (AGI) – will lead to one of two futures: a post-scarcity techno-utopia or the annihilation of humanity. For countless other species, the arrival of humans spelled doom. We weren't tougher, faster or stronger – just smarter and better coordinated. In many cases, extinction was an accidental byproduct of some other goal we had.

Zohran Mamdani gave Hinge an unofficial boost last month when the New York mayoral candidate revealed that he met his wife, Rama Duwaji, through swiping. "There is still hope on those dating apps," he said on the Bulwark podcast a week before his stunning victory in the Democratic primary. The tidbit spread over social media, cementing the 33-year-old democratic socialist's status as a millennial everyman. A subsequent Cosmopolitan headline read: "Zohran Mamdani could make history (as the first NYC mayor to meet his wife on Hinge)." Representatives for Hinge would not comment, but plenty of eligible New Yorkers did, claiming they would redownload the app due to Mamdani's success, in spite of their dating fatigue.

As a child, Julian Ako would visit his maternal great-grandfather's home near Pilaa Beach in Kauai, Hawaii, where he and his family would gather edible fungi that grow on kukui trees and collect seaweed and fish from the reef. For about a decade, that land has belonged to Meta CEO Mark Zuckerberg, who is constructing a massive compound at an estimated cost that exceeds 300 million. WIRED can now reveal that Zuckerberg's property is atop a burial site: Ako's great-grandmother and her brother were buried on the land. After months of discussions with a Zuckerberg representative, Ako was successfully able to gain access to the property and identify and register the graves with the state Department of Land and Natural Resources, though he was not able to locate remains of other ancestors, who he believes could be buried on the property. In a report shared with WIRED, the state agency also confirmed "the probability (based on oral testimony) of additional burial sites."

Drones have rapidly evolved from backyard novelties into critical components of today's infrastructure and now, into one of the fastest-growing threats to our national security. As the CEO of one of the nation's largest drone technology companies and a former naval officer, I've seen firsthand how powerful these tools can be. I've also seen how dangerous they are when left unregulated; they become liabilities, capable of disruption, destruction and danger. Just days ago, amid deadly flash floods in Texas, a private drone collided with a rescue helicopter during an active life-saving mission. The crash forced the crew to land, grounding a critical asset in the middle of an unfolding emergency.

While no one knows when or where the next major war will break out, what is becoming clear is that next time the United States engages directly in a conflict, U.S. combat units will be sharing their battle space with a different type of force -- drones, lots of them. In a push for the world's most powerful military to "meet the demands of 21st-century warfighting," Defense Secretary Pete Hegseth has ordered the Pentagon to fast-track the adoption and boost the number of various small drones deployed across the force, treating them as "consumable or expendable" capabilities similar to bullets, hand grenades and other munitions. The new initiative aims to ramp up the production, experimentation and fielding of small unmanned systems weighing less than 55 pounds (25 kilograms). This includes one-way, "kamikaze" attack drones and loitering munitions to maintain "battlefield superiority" as Washington's geopolitical and technological rivalry with Beijing intensifies.

In 2023 Elon Musk launched Grok, an AI chatbot marketed as providing "unfiltered answers" on X. In part, it was reportedly created to counter other machines that Musk saw as being trained to be "politically correct". Fast forward to 2025 and Grok is no stranger to controversy – sharing antisemitic content and white genocide conspiracy theories, and referring to itself as MechaHitler. One X user, Will Stancil, has even been the subject of extreme, violent, and individually tailored assault fantasies created by Grok, as he tells Nosheen Iqbal. "It's alarming and you don't feel completely safe when you see this sort of thing," he says.

Causal forests have become a popular tool for estimating how treatment effects vary across individuals (Wager and Athey, 2018). They are used in a growing number of domains--including marketing, operations, economics, and public policy--to personalize interventions and inform targeting strategies. Since 2019, dozens of papers in INFORMS journals alone have applied causal forests to experimental or observational data (see Appendix C), often with the goal of estimating individual-level treatment effects. The method builds on a familiar idea: instead of estimating a single average effect for the whole population, we split the population into subgroups based on observed features and estimate effects within each group. This is conceptually similar to how random forests estimate outcomes, except now the goal is to estimate causal effects. But there is a crucial modeling difference: unlike random forests, which typically use the full training data for both splitting and estimation, causal forests often divide the training data in two--using one part to decide how to form the subgroups, and the other to estimate effects within them. This practice, known as honest estimation, is meant to prevent overfitting and selection bias (Athey and Imbens, 2016). It is the default in widely used software packages such as grf (Athey et al., 2019) and EconML (Battocchi et al., 2019), and is commonly recommended in applied research. But is this default always a good idea? 1

Bridge maintenance and safety are essential for transportation authorities, and Non-Destructive Evaluation (NDE) techniques are critical to assessing structural integrity. However, interpreting NDE data can be time-consuming and requires expertise, potentially delaying decision-making. Recent advancements in Large Language Models (LLMs) offer new ways to automate and improve this analysis. This pilot study introduces a holistic assessment of LLM capabilities for interpreting NDE contour maps and demonstrates the effectiveness of LLMs in providing detailed bridge condition analyses. It establishes a framework for integrating LLMs into bridge inspection workflows, indicating that LLM-assisted analysis can enhance efficiency without compromising accuracy. In this study, several LLMs are explored with prompts specifically designed to enhance the quality of image descriptions, which are applied to interpret five different NDE contour maps obtained through technologies for assessing bridge conditions. Each LLM model is evaluated based on its ability to produce detailed descriptions, identify defects, provide actionable recommendations, and demonstrate overall accuracy. The research indicates that four of the nine models provide better image descriptions, effectively covering a wide range of topics related to the bridge's condition. The outputs from these four models are summarized using five different LLMs to form a comprehensive overview of the bridge. Notably, LLMs ChatGPT-4 and Claude 3.5 Sonnet generate more effective summaries. The findings suggest that LLMs have the potential to significantly improve efficiency and accuracy. This pilot study presents an innovative approach that leverages LLMs for image captioning in parallel and summarization, enabling faster decision-making in bridge maintenance and enhancing infrastructure management and safety assessments.

Coherent X-ray scattering techniques are critical for investigating the fundamental structural properties of materials at the nanoscale. While advancements have made these experiments more accessible, real-time analysis remains a significant bottleneck, often hindered by artifacts and computational demands. In scanning X-ray nanodiffraction microscopy, which is widely used to spatially resolve structural heterogeneities, this challenge is compounded by the convolution of the divergent beam with the sample's local structure. To address this, we introduce DONUT (Diffraction with Optics for Nanobeam by Unsupervised Training), a physics-aware neural network designed for the rapid and automated analysis of nanobeam diffraction data. By incorporating a differentiable geometric diffraction model directly into its architecture, DONUT learns to predict crystal lattice strain and orientation in real-time. Crucially, this is achieved without reliance on labeled datasets or pre-training, overcoming a fundamental limitation for supervised machine learning in X-ray science. We demonstrate experimentally that DONUT accurately extracts all features within the data over 200 times more efficiently than conventional fitting methods.