After Ukraine launched a sudden drone assault on Russian installations, it brought new attention to the U.S.' own vulnerabilities, regardless of which side the U.S. stood on Kyiv's attack. In recent years, Chinese Communist Party-linked entities have commercially targeted land around the U.S., including in the vicinity of sensitive installations like the Grand Forks Air Force Base in North Dakota. The Fufeng Group's 300-acre farmland purchase in 2021 first raised the collective antennae of Congress to such under-the-radar transactions – and even Florida Gov. Ron DeSantis swiftly banned them in his state as a result, among other efforts around the country. On Tuesday, North Dakota's senators agreed that the U.S. must remain vigilant for any malign activity, whether it be from relatively novel drone assaults to potential espionage through real estate transactions. An explosion of a drone is seen in the sky over the city during a Russian drone strike, amid Russia's attack on Ukraine, in Kyiv, Ukraine, on April 24, 2025.

The role of natural language constraints in Safe Multi-agent Reinforcement Learning (MARL) is crucial, yet often overlooked. While Safe MARL has vast potential, especially in fields like robotics and autonomous vehicles, its full potential is limited by the need to define constraints in pre-designed mathematical terms, which requires extensive domain expertise and reinforcement learning knowledge, hindering its broader adoption. To address this limitation and make Safe MARL more accessible and adaptable, we propose a novel approach named Safe Multi-agent Reinforcement Learning with Natural Language constraints (SMALL). Our method leverages fine-tuned language models to interpret and process free-form textual constraints, converting them into semantic embeddings that capture the essence of prohibited states and behaviours. These embeddings are then integrated into the multi-agent policy learning process, enabling agents to learn policies that minimize constraint violations while optimizing rewards. To evaluate the effectiveness of SMALL, we introduce the LaMaSafe, a multi-task benchmark designed to assess the performance of multiple agents in adhering to natural language constraints. Empirical evaluations across various environments demonstrate that SMALL achieves comparable rewards and significantly fewer constraint violations, highlighting its effectiveness in understanding and enforcing natural language constraints.

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. Pope Francis issued a warning on artificial intelligence Tuesday, urging those behind the technology to "be vigilant" during their work. The Pope made the statement in his message marking New Year's Day, which the Vatican traditionally releases far in advance. Francis, 86, has joked in the past that he is far from technologically savvy, but said Tuesday that AI must be used in a "responsible way." "Pope Francis calls for an open dialogue on the meaning of these new technologies, endowed with disruptive possibilities and ambivalent effects. He recalls the need to be vigilant and to work so that a logic of violence and discrimination does not take root in the production and use of such devices, at the expense of the most fragile and excluded," the message read.

Gatestone Institute senior fellow Gordon Chang weighs in on Treasury Secretary Janet Yellen's upcoming trip to China and the increase in Chinese nationals at the southern border on'The Ingraham Angle.' China has significantly expanded its legal framework to target those expected to or affiliated with threatening national security, putting pressure on the relationship between foreigners in China and Chinese working with foreign entities across all fields. Adding pressure to the already fragile relations, Chinese citizens are called upon to be vigilant against espionage and national security risks as part of a broader whole-of-society approach. The amendment is one of the latest attempts by Chinese lawmakers to control the flow of information among growing national security concerns. Recently, authorities closed its most extensive academic database, the privately owned China National Knowledge Infrastructure for several non-Chinese institutes, also the country's financial database restricted foreign access.

A bipartisan panel of voters weighed in on the future of artificial intelligence and growing concerns surrounding the potential dangers of the emerging technology. Imagine waking up tomorrow to discover that MLB commissioner Rob Manfeld made another rule change to America's pastime: All performance-enhancing drugs are legal. Imagine the increases in strength, speed, and stamina that baseball would have to contend with as players from every team experimented with pharmaceutical cocktails to give them the ultimate advantage. What we're about to see in the 2024 election cycle, with the introduction of artificial intelligence, is the rise of "performance-enhancing digital." AI will revolutionize politics in the months ahead.

How can humans stay in control of advanced artificial intelligence systems? One proposal is corrigibility, which requires the agent to follow the instructions of a human overseer, without inappropriately influencing them. In this paper, we formally define a variant of corrigibility called shutdown instructability, and show that it implies appropriate shutdown behavior, retention of human autonomy, and avoidance of user harm. We also analyse the related concepts of non-obstruction and shutdown alignment, three previously proposed algorithms for human control, and one new algorithm.

It was 7 a.m. when Benjamin Perkin's parents got an alarming call. The person on the other end of the line identified themselves as a lawyer and said Perkin had killed a U.S. diplomat in a car accident, was in jail, and needed money for legal fees. The lawyer then put Perkin on the phone. "Hey Mom and Dad, I love you. I need this money," he said.

She was speaking after SingularityNET unveiled Grace, a new robot nursing assistant at the 14th Conference on Artificial General Intelligence …

Those at risk are always vigilant for the signs of a stroke in progress, but no one can be vigilant when they're sleeping, meaning thousands of people suffer "wake-up strokes" that are only identified hours after the fact. Zeit Medical's brain-monitoring wearable could help raise the alarm and get people to the hospital fast enough to mitigate the stroke's damage and potentially save lives. A few decades ago, there wasn't much anyone could do to help a stroke victim. But an effective medication entered use in the '90s, and a little later a surgical procedure was also pioneered -- but both need to be administered within a few hours of the stroke's onset. Orestis Vardoulis and Urs Naber started Zeit ("time") after seeing the resources being put towards reducing the delay between a 911 call regarding a stroke and the victim getting the therapy needed.

AI and machine learning have been driving significant transformations across virtually all industries over the last few years. By taking advantage of the technology, businesses can make sense of vast amounts of data to drive intelligent decision making and create new, innovative services to improve customer experiences. The banking sector in particular has benefited from AI and machine learning, especially when it comes to fighting fraud, which is a continuous, ever-changing threat that banks need to remain vigilant to. At the same time, AI and machine learning powered technologies also ensure that security infrastructure doesn't compromise on the customer experience. For example, consumers want to stay safe, without having to deal with unnecessary false positives or blocked transactions.