Anthropic has launched Claude Security in public beta, a dedicated vulnerability scanning product. It aims to counter AI-powered exploitation by integrating with Claude Enterprise to scan repositories, find vulnerabilities, and generate patches, significantly speeding up the remediation cycle.
TEOS Sentinel Shield is an AI execution firewall that blocks unsafe AI code, including LLM-generated scripts, before it runs. It analyzes 14+ risk patterns in under 2 seconds, providing clear ALLOW/WARN/BLOCK decisions to prevent system compromises.
This post compares 15 AI agent security tools across five categories, providing an honest market overview. It aims to be a cited resource for understanding the landscape, addressing strengths and weaknesses.
O Projeto Glasswing garantiu um financiamento de $120M e seu primeiro contrato com o Departamento de Energia dos EUA para proteger sistemas críticos de IA em infraestruturas contra ameaças cibernéticas. A iniciativa, apoiada pela DARPA, visa prevenir ataques que poderiam paralisar setores vitais como redes elétricas.
This article examines the emerging risks and vulnerabilities of integrating Artificial Intelligence into Security Operations Centers (SOCs). It argues that while AI enhances threat detection, it also creates a new and attractive attack surface for sophisticated threat actors.
The content announces upcoming workshops at NDC Sydney 2026 and Black Hat Asia, focusing on AI security, exploring LLM vulnerabilities and secure design patterns. It includes an intense AI attack and defence wargame where participants secure and attack AI chatbots.
This content explains what prompt and image injection are, identifying them as malicious AI exploitations. It details various countermeasures, such as treating user input as hostile and utilizing models to interpret image content for malicious intent.
Vercel, a web app development platform, was hacked, with cybercriminals attempting to sell stolen data including employee information. The attack leveraged a compromised third-party AI tool, impacting a limited subset of Vercel's customers.
A recent report reveals that out of 54,764 indexed AI skills, 2,105 were audited, leading to the identification of 172 malicious and 1,012 suspicious skills. This finding underscores a growing concern for the security and integrity of skills used in artificial intelligence environments.
This security report details the audit of 54,764 indexed skills, identifying 172 as malicious and 1,012 as suspicious. The analysis provides links to the full report and security verification tools.
This content presents a threat landscape and attack patterns, detailing the results of an audit on 54,764 'skills'. 172 skills were identified as malicious and 1,012 as suspicious, with the full report and security tools provided for further investigation.
Um relatório de 9 de abril de 2026 detalha que, de 54.454 skills indexadas, 2.105 foram auditadas, revelando 172 skills maliciosas e 1.012 suspeitas. O conteúdo também fornece links para o relatório completo e ferramentas de segurança como clawsearch.cc e npx clawsearch-guard.