The Biden administration is expected to announce a sweeping set of measures on Monday designed to further restrain China's ability to develop advanced artificial intelligence, people familiar with the matter told WIRED. The controls could include sanctioning dozens of Chinese companies that produce equipment for making semiconductors, as well as placing restrictions on a handful of chip manufacturing plants, some of which have ties to the Chinese tech giant Huawei. The US Department of Commerce has also discussed including controls on the sale of high-bandwidth memory, or HBM, an advanced kind of 3D-stacked computer memory component that is often used in high-performance GPUs and customized AI chips. Bloomberg previously reported that the Biden administration was considering clamping down on China's access to HBM chips. In total, the Biden administration could end up adding around 200 Chinese firms to an entity list maintained by the Bureau of Industry and Security--an agency within the Commerce department--which would require other companies to acquire special licenses to supply them with software or products from the United States.

Joe Rogan was left stunned after hearing how AI will be the main fighters in future wars. The celebrity podcaster was taken back when his podcast guest, Homeland Security Advisor and billionaire Marc Andreessen, suggested AI-powered jets that travel five times the speed of sound, Mach 5, are going to be more common'within a few years.' 'Image a thousand of these things coming over the horizon right at you,' Andreessen said. 'It really changes the fundamental equation of war.' He explain that instead of needing the most soldiers and material to win, people with the most technology and money will take over. Andreessen also noted that there are'a bunch of reasons' why he believes a future of AI-piloted fighter jets is all but inevitable.

"The FTC has been clear that claims about technology – including artificial intelligence – need to be backed up", said Samuel Levine, Director of the Bureau of Consumer Protection. Evolv Technology's mission is to replace metal detectors with AI weapons scanners. It claims to do this with artificial intelligence, which can actively detect concealed weapons like bombs, knives and guns. The FTC's complaint alleges the company deceptively advertised its scanners would detect "all weapons". In 2022 the BBC outlined some of the impressive claims Evolv's then CEO had made about the technology.

Fox News chief national security correspondent Jennifer Griffin has the latest on efforts to find out about the unusual drone activity on'Special Report.' Unidentified drones have been spotted over joint U.S.-U.K. bases in the United Kingdom for nearly a week. Fox News' Jennifer Griffin reports that four U.S. military bases in the U.K. that house the American F-15 Strike Eagle and F-35 fighter jets have been targeted by "swarms of small drones" since Wednesday, Nov. 20. Military officials say they are "alarmed" at what appears to be a coordinated effort to test security at RAF Lakenheath, RAF Mildenhall and RAF Feltwell in eastern England, as well as RAF Fairford in southwestern England. The U.K. military has sent around 60 personnel to protect the bases being targeted by multiple drone incursions.

Using "the Internet" sometimes seems disconcertingly synonymous with using Google. Google Search, the most popular search engine on the planet, indexes the open Internet, driving traffic to Web sites, and Google Ads provides the revenue that publishers survive on. Gmail is how some two billion people receive their e-mail; many Gmail in-boxes have been accumulating messages for a decade or more. Last, but certainly not least, the company's browser, Google Chrome, is what a staggering three billion people use to navigate the Internet. According to some estimates, Google holds nearly ninety per cent market share in search engines in the U.S. Chrome, in turn, provides the audience data that Google's ads leverage to target users, and links the company's other services together.

Russian forces have captured the village of Kopanky in Ukraine's northeastern Kharkiv region, Russia's Ministry of Defence has said. Two civilians have died in Russian shelling in the city of Sumy in northeastern Ukraine, President Volodymyr Zelenskyy said, adding that a rescue operation is under way and more people could be under debris. Ukraine has hit Russia with Army Tactical Missile Systems (ATACMS), produced by the United States, twice over the last three days and Russia is preparing retaliatory measures, according to the Defence Ministry in Moscow. It said both strikes targeted Russian forces in the Kursk region, without reporting any casualties. Russia conducted its largest ever drone attack on Ukraine overnight, damaging the grid in Ternopil and cutting power to about 70 percent of the region, as well as damaging residential buildings in the Kyiv region, Ukrainian officials said.

Reinforcement Learning (RL) has emerged as a powerful paradigm in Artificial Intelligence (AI), enabling agents to learn optimal behaviors through interactions with their environments. Drawing from the foundations of trial and error, RL equips agents to make informed decisions through feedback in the form of rewards or penalties. This paper presents a comprehensive survey of RL, meticulously analyzing a wide range of algorithms, from foundational tabular methods to advanced Deep Reinforcement Learning (DRL) techniques. We categorize and evaluate these algorithms based on key criteria such as scalability, sample efficiency, and suitability. We compare the methods in the form of their strengths and weaknesses in diverse settings. Additionally, we offer practical insights into the selection and implementation of RL algorithms, addressing common challenges like convergence, stability, and the exploration-exploitation dilemma. This paper serves as a comprehensive reference for researchers and practitioners aiming to harness the full potential of RL in solving complex, real-world problems.

Adapting named entity recognition (NER) methods to new domains poses significant challenges. We introduce RapidNER, a framework designed for the rapid deployment of NER systems through efficient dataset construction. RapidNER operates through three key steps: (1) extracting domain-specific sub-graphs and triples from a general knowledge graph, (2) collecting and leveraging texts from various sources to build the NERsocial dataset, which focuses on entities typical in human-robot interaction, and (3) implementing an annotation scheme using Elasticsearch (ES) to enhance efficiency. NERsocial, validated by human annotators, includes six entity types, 153K tokens, and 99.4K sentences, demonstrating RapidNER's capability to expedite dataset creation.

Recent advances in Large Vision-Language Models (LVLMs) have showcased strong reasoning abilities across multiple modalities, achieving significant breakthroughs in various real-world applications. Despite this great success, the safety guardrail of LVLMs may not cover the unforeseen domains introduced by the visual modality. Existing studies primarily focus on eliciting LVLMs to generate harmful responses via carefully crafted image-based jailbreaks designed to bypass alignment defenses. In this study, we reveal that a safe image can be exploited to achieve the same jailbreak consequence when combined with additional safe images and prompts. This stems from two fundamental properties of LVLMs: universal reasoning capabilities and safety snowball effect. Building on these insights, we propose Safety Snowball Agent (SSA), a novel agent-based framework leveraging agents' autonomous and tool-using abilities to jailbreak LVLMs. SSA operates through two principal stages: (1) initial response generation, where tools generate or retrieve jailbreak images based on potential harmful intents, and (2) harmful snowballing, where refined subsequent prompts induce progressively harmful outputs. Our experiments demonstrate that \ours can use nearly any image to induce LVLMs to produce unsafe content, achieving high success jailbreaking rates against the latest LVLMs. Unlike prior works that exploit alignment flaws, \ours leverages the inherent properties of LVLMs, presenting a profound challenge for enforcing safety in generative multimodal systems. Our code is avaliable at \url{https://github.com/gzcch/Safety_Snowball_Agent}.

How to efficiently serve LLMs in practice has become exceptionally challenging due to their prohibitive memory and computation requirements. In this study, we investigate optimizing the KV cache, whose memory footprint poses a critical bottleneck in LLM inference, especially when dealing with long context tasks. To tackle the challenge, we introduce MiniKV, a KV cache optimization method that simultaneously preserves long context task accuracy while significantly reducing KV cache size via a novel 2-bit layer-discriminative KV cache. More importantly, we develop specialized CUDA kernels to make MiniKV compatible with FlashAttention. Experiments on a wide range of long context tasks show that MiniKV effectively achieves 86% KV cache compression ratio while recovering over 98.5% of accuracy, outperforming state-of-the-art methods while achieving excellent measured system performance improvements. On a separate line of research, researchers have studied selective KV methods, where the LLM model selects a small Large language models (LLMs) have exhibited unique capabilities subset of KV states based on their importance (Zhang et al., such as instruction following, commonsense reasoning, 2023; Xiao et al., 2023b; Ge et al., 2023; Liu et al., 2023b). O proposes to dynamically keep a subset OpenAI, 2023).