Tag: LLM
-
Sweet Security launches LLM-powered cloud detection engine
First seen on scworld.com Jump to article: www.scworld.com/brief/sweet-security-launches-llm-powered-cloud-detection-engine
-
How organizations can secure their AI code
Tags: ai, application-security, awareness, backdoor, breach, business, chatgpt, ciso, compliance, control, credentials, crime, cybersecurity, data, data-breach, finance, github, healthcare, LLM, malicious, ml, open-source, organized, programming, risk, risk-management, software, startup, strategy, supply-chain, technology, tool, training, vulnerabilityIn 2023, the team at data extraction startup Reworkd was under tight deadlines. Investors pressured them to monetize the platform, and they needed to migrate everything from Next.js to Python/FastAPI. To speed things up, the team decided to turn to ChatGPT to do some of the work. The AI-generated code appeared to function, so they…
-
OpenAI’s ChatGPT crawler can be tricked into DDoSing sites, answering your queries
The S in LLM stands for Security First seen on theregister.com Jump to article: www.theregister.com/2025/01/19/openais_chatgpt_crawler_vulnerability/
-
A Brief Guide for Dealing with ‘Humanless SOC’ Idiots
image by Meta.AI lampooning humanless SOC My former “colleagues” have written several serious pieces of research about why a SOC without humans will never happen (“Predict 2025: There Will Never Be an Autonomous SOC”, “The “Autonomous SOC” Is A Pipe Dream”, “Stop Trying To Take Humans Out Of Security Operations”). But I wanted to write…
-
Cisco Unveils AI Defense to Stand Against Model Safety Risks
Product Head Jeetu Patel on How AI Defense Ensures Secure LLM Operations at Runtime. Cisco’s AI Defense platform addresses emerging safety and security risks in AI. By leveraging insights from Robust Intelligence, it offers model validation, threat prevention and integrated guardrails to protect against evolving challenges such as hallucinations and prompt injection attacks. First seen…
-
OWASP’s New LLM Top 10 Shows Emerging AI Threats
Ultimately, there is no replacement for an intuitive, security-focused developer working with the critical thinking required to drive down the risk of both AI and human error. First seen on darkreading.com Jump to article: www.darkreading.com/vulnerabilities-threats/owasps-llm-top-10-shows-emerging-ai-threats
-
Sweet Security Introduces Patent-Pending LLM-Powered Detection Engine, Reducing Cloud Detection Noise to 0.04%
Tel Aviv, Israel, 15th January 2025, CyberNewsWire First seen on hackread.com Jump to article: hackread.com/sweet-security-introduces-patent-pending-llm-powered-detection-engine-reducing-cloud-detection-noise-to-0-04/
-
Sweet Security Leverages LLM to Improve Cloud Security
Sweet Security today added a cloud detection engine to its cybersecurity portfolio that makes use of a large language model (LLM) to identify potential threats in real-time. First seen on securityboulevard.com Jump to article: securityboulevard.com/2025/01/sweet-security-leverages-llm-to-improve-cloud-security/
-
How to create de-identified embeddings with Tonic Textual Pinecone
To protect private information stored in text embeddings, it’s essential to de-identify the text before embedding and storing it in a vector database. In this article, we’ll demonstrate how to de-identify and chunk text using Tonic Textual, and then easily embed these chunks and store the data in a Pinecone vector database to use for…
-
SOAR buyer’s guide: 11 security orchestration, automation, and response products, and how to choose
Tags: access, ai, api, attack, automation, business, chatgpt, cisco, cloud, compliance, container, cybersecurity, data, detection, edr, endpoint, firewall, fortinet, gartner, google, group, guide, Hardware, ibm, incident response, infrastructure, intelligence, jobs, LLM, malware, mandiant, marketplace, microsoft, mitigation, monitoring, network, okta, risk, saas, security-incident, service, siem, soar, soc, software, technology, threat, tool, training, vulnerability, vulnerability-management, zero-daySecurity orchestration, automation, and response (SOAR) has undergone a major transformation in the past few years. Features in each of the words in its description that were once exclusive to SOAR have bled into other tools. For example, responses can be found now in endpoint detection and response (EDR) tools. Orchestration is now a joint…
-
Forscher: KI sorgt für effektiveres Phishing
Wie wirksam ist per LLM automatisch erzeugtes Phishing? Es ist gleichauf mit menschlich erzeugtem Spear-Phishing, sagen Forscher. First seen on heise.de Jump to article: www.heise.de/news/Forscher-KI-sorgt-fuer-effektiveres-Phishing-10232370.html
-
Sophos stellt Sprachmodell-Tool zur Verfügung – Tuning-Tool für LLMs als Open-Source-Programm
First seen on security-insider.de Jump to article: www.security-insider.de/sophosai-open-source-tool-large-language-models-a-7f503f54ce6f32d4c318a41e873e2a54/
-
Gen AI is transforming the cyber threat landscape by democratizing vulnerability hunting
Tags: ai, api, apt, attack, bug-bounty, business, chatgpt, cloud, computing, conference, credentials, cve, cyber, cybercrime, cyberespionage, cybersecurity, data, defense, detection, email, exploit, finance, firewall, flaw, framework, github, government, group, guide, hacker, hacking, incident response, injection, LLM, malicious, microsoft, open-source, openai, penetration-testing, programming, rce, RedTeam, remote-code-execution, service, skills, software, sql, tactics, threat, tool, training, update, vulnerability, waf, zero-dayGenerative AI has had a significant impact on a wide variety of business processes, optimizing and accelerating workflows and in some cases reducing baselines for expertise.Add vulnerability hunting to that list, as large language models (LLMs) are proving to be valuable tools in assisting hackers, both good and bad, in discovering software vulnerabilities and writing…
-
Will AI Code Generators Overcome Their Insecurities This Year?
In just two years, LLMs have become standard for developers, and non-developers, to generate code, but companies still need to improve security processes to reduce software vulnerabilities. First seen on darkreading.com Jump to article: www.darkreading.com/application-security/will-ai-code-generators-overcome-their-insecurities-2025
-
Garak An Open Source LLM Vulnerability Scanner for AI Red-Teaming
Garak is a free, open-source tool specifically designed to test the robustness and reliability of Large Language Models (LLMs). Inspired by utilities like Nmap or Metasploit, Garak identifies potential weak points in LLMs by probing for issues such as hallucinations, data leakage, prompt injections, toxicity, jailbreak effectiveness, and misinformation propagation. This guide covers everything you…
-
Im Kontext der Datensicherheit sind LLMs als Menschen zu betrachten
Große Sprachmodelle und die Frage der Data Security Sicherheitsfragen rund um LLMs. Angesichts der rasanten KI-Entwicklung wird immer deutlicher, dass die grundlegenden Leitplanken, Plausibilitätsprüfungen und prompt-basierten Sicherheitsmaßnahmen, die derzeit gelten, durchlässig und unzureichend sind. Bei der Entwicklung von Strategien zur Verbesserung der Datensicherheit in KI-Workloads ist es entscheidend, die Perspektive zu ändern und… First seen…
-
New LLM jailbreak uses models’ evaluation skills against them
First seen on scworld.com Jump to article: www.scworld.com/news/new-llm-jailbreak-uses-models-evaluation-skills-against-them
-
‘Bad Likert Judge’ Jailbreak Bypasses Guardrails of OpenAI, Other Top LLMs
A novel technique to stump artificial intelligence (AI) text-based systems increases the likelihood of a successful cyberattack by 60%. First seen on darkreading.com Jump to article: www.darkreading.com/cyberattacks-data-breaches/bad-likert-judge-jailbreak-bypasses-guardrails-openai-other-llms
-
LLMs Crack the Code: 95% Success Rate in Hacking Challenge
A recent study demonstrates the transformative potential of large language models (LLMs) in offensive cybersecurity tasks. Researchers Rustem Turtayev, Artem Petrov, Dmitrii Volkov, and Denis Volk have achieved a record-breaking... First seen on securityonline.info Jump to article: securityonline.info/llms-crack-the-code-95-success-rate-in-hacking-challenge/
-
SECURITY AFFAIRS MALWARE NEWSLETTER ROUND 26
Security Affairs Malware newsletter includes a collection of the best articles and research on malware in the international landscape. Now You See Me, Now You Don’t: Using LLMs to Obfuscate Malicious JavaScript Analyzing Malicious Intent in Python Code: A Case Study DigiEver Fix That IoT Thing! Botnets Continue to Target Aging D-Link Vulnerabilities OtterCookie, […]…
-
AI Could Generate 10,000 Malware Variants, Evading Detection in 88% of Case
Cybersecurity researchers have found that it’s possible to use large language models (LLMs) to generate new variants of malicious JavaScript code at scale in a manner that can better evade detection.”Although LLMs struggle to create malware from scratch, criminals can easily use them to rewrite or obfuscate existing malware, making it harder to detect,” Palo…
-
Die 10 häufigsten LLM-Schwachstellen
Tags: access, ai, api, application-security, awareness, breach, cloud, control, cyberattack, data, detection, dos, encryption, injection, least-privilege, LLM, ml, monitoring, privacy, RedTeam, remote-code-execution, risk, service, tool, update, vulnerability, zero-trust -
Sophos stellt Tuning-Tool für große Sprachmodelle als Open-Source-Programm zur Verfügung
Große Sprachmodelle (Large-Language-Models, LLMs) haben das Potenzial, die Arbeitslast zu automatisieren und zu reduzieren, einschließlich der von Cybersicherheitsanalysten und Incident-Respondern. Generischen LLMs fehlt jedoch das domänenspezifische Wissen, um diese Aufgaben gut zu bewältigen. Auch wenn sie mit Trainingsdaten erstellt wurden, die Cybersicherheitsressourcen enthalten, reicht dies oft nicht aus, um spezialisiertere Aufgaben zu übernehmen, die aktuelles…
-
Sophos stellt neues Trainings-Framework zur Optimierung der LLMs zur Verfügung
Durch den Einsatz von DeepSpeed wird die Skalierung großer Trainingsaufgaben ermöglicht, unter anderem durch parallele Datenverarbeitung. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/sophos-stellt-neues-trainingsframework-zur-optimierung-der-llms-zur-verfuegung/a39320/