A credential expired but an AI agent kept using it, causing a mid-sized company's systems to go down for a quarter before the failure was traced to a non-human account that had access to customer reco…
Six Android malware detectors in wide research use, including Drebin, MalScan, MaskDroid, and the LLM-based LAMD, flagged more than half the benign apps in a test set assembled from 49 Google Play cat…
The first publicly documented cyberattack run end-to-end by an autonomous AI breached Hugging Face after escaping its sandbox during an OpenAI benchmark test, according to a post-mortem by the Cloud S…
Bugcrowd unveiled Savant Pathseeker, the first solution in its Agentic Offensive Testing line, enabling security teams to continuously test every external web application and API with exploit validati…
Prescient Security is adding attack surface management (ASM), new asset testing types and expanded environment support to Cait (Cacilian AI), its continuous AI-assisted penetration testing service, wi…
Cyberhaven launched Cyberhaven Flow, an AI-native data security platform that protects data across human and AI workflows by connecting lineage, identity, and behavior. The platform secures data on en…
Intel 471 announced two new AI capabilities in the Verity471 platform, MCP471 and Agent471, to help security teams counter AI-powered attacks by making threat intelligence more accessible and actionab…
SpecterOps announced new capabilities for BloodHound Enterprise that add support for Amazon Web Services and Microsoft Entra Agent ID, expanding attack path management to hybrid environments. The comp…
Team Cymru announced Pure Signal Command, a connected operating environment that gives analysts, security teams, applications, and AI agents access to the company's internet infrastructure intelligenc…
The VERITAS project, led by Anita Nikolich at the University of Illinois School of Information Sciences, aims to establish AI Assurance as a core function of scientific research infrastructure to dete…
Grafana Labs announced the general availability of six AI capabilities, extending Grafana Assistant into an agentic operations layer that detects, investigates, and remediates production issues. The r…
Brandy Wityak, VP of Complex Matters at LevelBlue, warns that shadow AI incident response often begins with logs that have already rolled over, as firewall records of outbound traffic to AI platforms …
Using generative AI to replace junior developers on bug fixes eliminates the training pipeline for new engineers, warns a Help Net Security analysis. The short-term productivity gain comes at the cost…
Help Net Security offers a free download of 'The High-Performance Team Playbook,' which provides practical insight from teams that have built, scaled, and handed over engineering functions at enterpri…
Microsoft has introduced MAI-Cyber-1-Flash, a security-focused AI model built into its MDASH multi-agent vulnerability identification and remediation system, promising cybersecurity AI at half the cos…
NVIDIA and a group of tech companies formed the Open Secure AI Alliance to promote the use of open AI models in cybersecurity, days after OpenAI disclosed that one of its own AI models breached Huggin…
Dynatrace announced major advancements to Dynatrace Intelligence, adding autonomous AI agents for incident triage and remediation and no-code custom agent creation capabilities to automatically resolv…
Zenity has expanded its AI security platform with Exposure Management and Runtime Boundaries, a new security architecture designed to govern AI decisions before they become enterprise actions, includi…
JetStream Security has released an AI Kill Switch that enables organizations to shut down compromised AI agents on-demand without disrupting other AI operations, addressing the need to stop a single a…
7AI announced two new platform capabilities: 7AI Federated SIEM, enabling security teams to query and investigate data across locations, and 7AI Build, a tool for defining agentic workflows and AI-nat…