Articles tagged with Cybersecurity
A new study by the University of Cambridge found that many AI bots lack basic safety disclosures, including transparency about their abilities and potential risks. The 'AI Agent Index' revealed a significant transparency gap, with only four out of thirty agents publishing formal safety and evaluation documents.
Research from UNSW Sydney and ANU found that people with average face-recognition ability performed only slightly better than chance in spotting AI-generated faces. Even super-recognisers, who excel at recognizing real human faces, were not immune to being fooled by highly realistic AI-generated images.
A new study by Binghamton University School of Management reveals that businesses can improve their performance when they openly discuss and address cybersecurity issues. The research found a positive correlation between cybersecurity readiness and firm performance, suggesting that proactive measures can lead to better outcomes.
The University of Tennessee and UCOR have strengthened their partnership to provide training and educational opportunities for students and workforce development in nuclear and criticality safety. The MOU renewal will include projects, research, and curriculum focused on these areas.
Researchers say tech companies need to adopt user-centered approaches to create authentication systems that are secure by design. They call for more inclusive and consistent methods, avoiding one-size-fits-all models that burden users with complex security rules.
New machine-learning tools significantly improve phishing site detection accuracy, surpassing 95% precision/recall. The study evaluates ten classifiers across three public datasets using URL, domain, and content features.
A new report by Anglia Ruskin University and the University of York identifies key weaknesses in the UK's food system and recommends interventions to prevent disruptions from escalating into a crisis. The study suggests increasing energy security, diversifying food value chains, and promoting varied diets to reduce risks.
Researchers investigate real-world blockchain security incidents to extract representative attack patterns and categorize them into six classes based on standard blockchain architecture layers. The study also evaluates current detection and defense strategies, highlighting their strengths and limitations.
The VERIFHE project aims to develop an open-source library for verifiable homomorphic encryption, enabling private and reliable inference. The ERC Proof of Concept grant will support the development of practical tools for machine learning services.
The Association for Computing Machinery has named 72 new Fellows for their technical innovations and service to the field. The honorees represent the top 1% of professionals worldwide, with contributions in AI for healthcare, computer graphics, data management, and more.
Researchers have identified a new vulnerability in AI-powered robots, where malicious text on signs or objects can hijack their decision-making. The study, led by UC Santa Cruz professors Alvaro Cardenas and Cihang Xie, presents the first academic exploration of environmental indirect prompt injection attacks against embodied AI systems.
A new approach uses multiple LLM agents collaborating through iterative interactions to improve AI safety checks, outperforming existing methods in F1-score and achieving better trade-offs between catching harmful and harmless prompts. The system divides work into two teams: generator and analyzer, with logs capturing past mistakes and...
Researchers found that nearly half of geostationary satellite communications are not encrypted, putting sensitive data such as cellular texts, voice calls, and military information at risk. The study led to immediate action from companies like T-Mobile, Walmart, and KPU Telecom, which enabled encryption in response to the findings.
New research from the University of East London reveals widespread online risk-taking among young people, with many engaging in illegal behavior before crossing a legal boundary. The study highlights clear pathways into cybercrime, often driven by risk, opportunity, and lack of understanding about online laws.
A new study reveals that young people are at risk of engaging in serious online offenses through gradual progression from low-risk behaviors like digital piracy. Researchers found that nearly half reported engaging in illegal online behavior early on, with many describing a pattern of repeat offending.
A recent paper highlights the need for defense mechanisms covering software, programs, and physical components of quantum computing systems. Key findings include the risk of crosstalk, intellectual property theft, and lack of end-to-end protection, emphasizing the need for safeguarding quantum computers from ground up.
Andreas Zeller recognized for his work on automated debugging and security testing, with a total of nine Test of Time Awards. He is one of the few researchers to receive two ERC Advanced Grants from the European Research Council.
The SHASAI project addresses cybersecurity risks in AI systems from design to real-world operation, combining secure hardware and software, risk-driven engineering, and real-world validation. The project will demonstrate its methods and tools in three diverse scenarios, supporting Europe's efforts to promote trustworthy AI.
The Association for Computing Machinery has named 61 new Distinguished Members, recognized for their contributions to AI for healthcare, data management, human-computer interaction, and other areas. The recipients include experts from top universities, corporations, and research institutions worldwide.
Rayna Dimitrova's SyReP project aims to eliminate the separation between decision-making and data in software development, enabling developers to specify system behavior and generate implementation automatically. The ERC Consolidator Grant will support her research on novel symbolic algorithms and realizability certificates.
Researchers analyzed 12 years of hospital website data and found pixel tracking in 66% of observations, significantly increasing breach risk. Hospitals using third-party pixels saw a 1.4 percentage point increase in breach probability compared to baseline rates.
A new study by Pius Ewoh argues that healthcare cybersecurity requires socio-technical solutions, not just technical fixes. The framework combines human behaviour monitoring, organisational learning, and intelligent incident-response capabilities to identify and prevent vulnerabilities.
A new study by Emmanuel Anti at the University of Vaasa suggests that empathetic security models can strengthen cybersecurity practices and prevent insider threats. The research proposes co-creating cybersecurity policies with employees, focusing on understanding their needs, motivations, and emotional well-being.
Bahaa Eltahawy's dissertation highlights the importance of integrated cybersecurity strategies for protecting critical infrastructure. The study reveals that current standards have limitations and proposes a model that expands protection into seven domains.
The partnership aims to develop solutions for emerging digital threats through joint research initiatives in cryptography, AI security, and blockchain technology. The collaboration will facilitate researcher mobility and identify opportunities for collaborative funding.
Researchers at SwRI discovered a security vulnerability in the Signal Level Attenuation Characterization (SLAC) protocol used in electric vehicle-to-grid communications. The team successfully modeled and replicated a machine-in-the-middle attack that could manipulate or halt EV charging processes.
A study by Griffith University researchers reveals that smartphone sharing is a growing issue, with 20% of identity theft perpetrators being current or former intimate partners. The study argues that one-user/one-device threat models are inadequate for addressing interpersonal cybersecurity risks.
Security researchers have developed a defense mechanism that protects against cryptanalytic parameter extraction attacks on AI systems. By making neurons in the same layer similar to each other, the defense creates a barrier of similarity that makes it difficult for attacks to proceed.
A University of Delaware team, funded by NSF, creates next-generation secure computer chips using AI and game theory. The project aims to predict attacks and set new benchmarks for global hardware security, enabling the design of safer and more trustworthy systems.
Researchers from the University of Florida discovered critical security vulnerabilities in portable genetic sequencers, exposing them to potential attacks and data breaches. The devices' reliance on insecure connections to laptops or unsecured networks amplifies these risks.
The University of Houston has been redesignated by the NSA as a Center for Academic Excellence in Cyber Research (CAE-R) through 2029. This recognition underscores UH's growing role in advancing research and workforce training to strengthen US resilience against evolving cyber threats.
A game theoretical model of cyberwarfare assesses Attacker and Defender costs, revealing that networks with fewer attack vectors benefit Defenders. Converging technological capabilities amplify conflicts, while centralized government control can provoke more aggressive attacks.
Doowon Kim, an assistant professor at the University of Tennessee, received a prestigious award from ACM CCS for his decade-long study on Advanced Persistent Threats (APTs). His research provides a comprehensive understanding of APT evolution and global patterns, shedding new light on these sophisticated cyberattacks.
Researchers found that providing feedback to everyone after the entire simulation ends leads to better learning outcomes. The study's findings suggest that companies can design more effective software tools using this approach, which has been implemented by KnowBe4.
A new security framework based on blockchain technology and distributed reinforcement learning ensures secure data storage and transmission while adapting to evolving threats. The framework demonstrated improved memory consumption and transaction latency compared to existing approaches.
Researchers are developing training-free detection methods and novel countermeasures to protect sensitive information from emerging wireless eavesdropping techniques. By manipulating wireless channels, these defenses can obscure or distort signal patterns that attackers rely on.
A new study by Yifan Yu offers guidance on how to deploy emotion AI in various scenarios, emphasizing the importance of balancing human involvement with AI's emotional detection capabilities. The analysis showed that emotion AI works best when integrated with human employees, and some scenarios are better handled by humans alone.
Researchers developed a cyberattack detection system that uses federated learning and cloud coordination to detect DDoS attacks in 6G-ready smart grids without exposing user energy-use data. The system achieved high accuracy and precision, but trade-offs were observed in terms of resources consumption.
Researchers at the University of Missouri have developed an AI-powered method to detect hidden hardware trojans in chip designs, offering a 97% accurate solution. The approach leverages large language models to scan for suspicious code and provides explanations for detected threats.
New WiFi technology can create images of individuals and surroundings by passively recording communication in radio networks, raising concerns about privacy. Researchers call for protective measures and privacy safeguards in the forthcoming IEEE 802.11bf WiFi standard.
The platform allows users to create personalized LLM-based chatbots, serving as a 24/7 teaching assistant or scraping campus websites to find needed resources. Illinois Chat has already been utilized in various departments and research groups, providing valuable feedback to make it more practical.
A new study reveals that multitasking makes phishing detection harder, but introducing brief reminders can significantly improve accuracy. Researchers found that working memory load and message framing cues affect detection performance.
Researchers at Texas A&M University have developed a new framework to protect industrial processing facilities from cyber threats. The framework identifies vulnerabilities, detects abnormal activity in real-time, and provides safeguards and mitigation strategies to maintain safe operations.
Researchers at NC State University identified a hardware vulnerability that allows attackers to compromise AI privacy by exploiting ML accelerator behavior. The GATEBLEED vulnerability exploits power gating in AI accelerators, creating an observable timing channel for attackers.
Researchers at the University of Sydney have developed end-to-end encryption for git services, protecting sensitive code from cybersecurity threats. The encryption aligns seamlessly with existing platforms like Github and Bitbucket, reducing computational resources needed while encrypting only edits.
Florida International University researchers have developed a defensive system called SHIELD that can detect and neutralize cyberattacks on drones in real time. The system allows the drone to finish its mission while providing robust recovery mechanisms, making it crucial for securing commercial drone use across industries.
Researchers Joseph Clark, Doowon Kim, and Joon Sue Lee received NSF CAREER awards for their groundbreaking work in chemistry, computer science, and physics. They are developing new methods to track pharmaceutical drugs and detect phishing websites, while studying the properties of quantum materials.
The researchers have successfully demonstrated a four-dimensional QKD system with high efficiency and low measurement error rates. This breakthrough enables secure data transmission over long distances, with potential applications in fields such as finance and government.
Liberate AI project develops an AI model capable of predicting long-term outcomes and potential complications in ischemic stroke patients. The model will be trained using Swarm Learning technology, aiming for explainability and transparency, while striking a balance between these features and accuracy.
The CIAS Community Cybersecurity Clinic will expand its reach to more organizations with the support of a $100,000 grant from Craig Newmark Philanthropies. The clinic provides low- and no-cost cybersecurity services to organizations in need, as well as hands-on workforce development opportunities for UT San Antonio students.
The European Lighthouse on Secure and Safe AI (ELSA) network has grown to include 41 members, expanding its expertise in pressing AI research topics and fostering knowledge exchange. The network's founding members are renowned for their work on Safety, Security, Artificial Intelligence, and Machine Learning.
Researchers at UBC have developed a new NFC-based gesture system that can distinguish between nine distinct gestures with high accuracy. This technology has the potential to improve accessibility for users with mobility issues and reduce the spread of germs through touchpads.
The partnership aims to provide cutting-edge professional development to public sector and critical infrastructure leaders across Canada, equipped with strategic skills to outthink and outmaneuver threats. The initiative strengthens Canada's cyber resilience by enhancing national cyber resilience.
A recent study found that cybersecurity training programs do little to prevent employees from falling for phishing scams. Researchers sent 10 different phishing email campaigns to over 19,500 employees at UC San Diego Health and found that embedded phishing training only reduced the likelihood of clicking on a phishing link by 2%. As t...
AI TechX, a UT Knoxville initiative, awards $60,000 seed funding to nine projects addressing real-world challenges through AI research and innovation. The funded projects accelerate adoption of AI across various disciplines, enabling high-impact collaborations and job creation.
Southwest Research Institute (SwRI) has installed a private 5G network to boost security, eliminate dependency on outside mobile networks, and expand off-site capabilities for advanced 5G projects. The secure network supports various 5G advancements, including cybersecurity research, drone technology, and antenna development.
The University of Texas at San Antonio has launched the College of AI, Cyber and Computing, which will serve over 5,000 students pursuing degrees in rapidly growing career fields. The college is positioned to advance technological education and research, with faculty securing major grants and launching industry partnerships.
The £250 million investment will create an Advanced Marine Technology Hub at the University of Plymouth, leveraging its expertise in autonomous marine systems, maritime cyber security, and renewable energy. This initiative aims to boost the city's economy and enhance UK's national resilience.
Researchers at UCR have developed a method to preserve AI safeguards in open-source models by retraining internal structure to detect and block dangerous prompts. The approach avoids external filters or software patches, instead changing the model's fundamental understanding of risky content.
The Wits MIND Institute has received a $1 million boost from Google.org, enabling it to drive next-generation breakthroughs in natural and artificial intelligence. The partnership aims to advance the scientific understanding of both natural and artificial intelligence, foster breakthrough research and technological innovation.