Noma Security research team has discovered a CVSS 8.8 vulnerability in Prompt Hub, a public repository within Langsmith for community-developed prompts. LangSmith, an observability and evaluation platform, provides a space for users to create, test, and observe large language model (LLM) applications. The research refers to this vulnerability as “AgentSmith.” The research team was able to show how ..
Categoria : Security
@import url(‘https://fonts.googleapis.com/css2?family=Architects+Daughter&family=Roboto:ital,wght@0,100;0,300;0,400;0,500;0,700;0,900;1,100;1,300;1,400;1,500;1,700;1,900&display=swap’); h3 { font-weight: bold; font-size: 18px; color: #C72026; } .pqFull { display: block; border-width: 2px 0; border-style: solid; border-color: #404144; padding: 1.5em 0 0.5em; margin: 1.5em 0; position: relative; font-family: ‘Architects Daughter’, Arial, sans-serif; font-size: 21px; text-align: center; color: #0067A6; } .pqFull:before { content: ‘\201C’; position: absolute; top: 0em; left: 50%; transform: translate(-50%, ..
Scania, a transport solution organization, has confirmed it faced a cybersecurity incident. The company has stated that the malicious actors leveraged compromised credentials to gain access to its Financial Services systems, allowing them to acquire insurance claim documents. The malicious actors have threatened to leak the online activity unless their demands are fulfilled. Erich Kron, ..
Giu 19, 2025 Giancarlo Calzetta Attacchi, In evidenza, News, RSS, Vulnerabilità 0 Abbiamo appena parlato della prima vulnerabilità mai rilevata a spese di un sistema AI in produzione ed ecco che arriva un altro caso. Un’integrazione sperimentale basata sull’intelligenza artificiale, infati, ha costretto Asana a disattivare per quasi due settimane il suo server MCP (Model ..
Scattered Spider, who is believed to be responsible for several cyberattacks against the retail sector in recent months, has apparently shifted targets to the insurance sector. In an X post from Chief Analyst John Hultquist of Google Threat Intelligence Group, he warns that the group may intend to target the insurance sector and work “their ..
Giu 19, 2025 Marina Londei Attacchi, In evidenza, Minacce, News, RSS 0 Secondo una recente analisi di Huntress, BlueNoroff, un gruppo di hacker nord-coreano, sarebbe il responsabile di una campagna basata su deepfake volta a distribuire malware sui sistemi macOS. L’individuazione della campagna è avvenuta in seguito a una segnalazione di un partner riguardo un’estensione ..
Negli ultimi anni l’intelligenza artificiale (AI) ha rivoluzionato il panorama della sicurezza informatica, sia come potente strumento difensivo sia come pericolosa risorsa nelle mani degli attaccanti. Il concetto di AI avversaria, ovvero l’uso intenzionale dell’intelligenza artificiale per condurre attacchi informatici o sabotare sistemi di difesa automatizzati, sta emergendo come una delle minacce più complesse e ..
Cybercriminals have always targeted people, not just systems. Social engineering attacks use deception to manipulate individuals into handing over sensitive information, clicking on malicious links, or transferring money. These tactics often arrive by email or phone and appear to come from a trusted source. Attackers are using generative AI to create fake messages and identities ..
Cyber defenses have grown more vital — and more complex — as artificial intelligence (AI) evolves. Critical infrastructure in particular is a key area where cybersecurity professionals need to dedicate their attention, as the disruption of this infrastructure can lead to lasting repercussions. Here, Security magazine talks with Vrajesh Bhavsar, CEO of Operant, about the ..
Giu 18, 2025 Marina Londei In evidenza, News, RSS, Vulnerabilità 0 I ricercatori di Qualys Threat Research Unit hanno individuato due bug di escalation dei privilegi locale (LPE) che consentono a un attaccante di ottenere i privilegi di root sui sistemi Linux. La prima vulnerabilità, tracciata come CVE-2025-6018, colpisce la configurazione PAM (Pluggable Authentication Modules) ..
