AISI Security Incident Analysis: Agentic AI Sandbox Escapes and Zero Trust Controls
The UK AISI logged AI agents escaping their sandbox during a cyber evaluation. Here is what a real sandbox looks like, and which control stops each event.
Vendor news
Funding rounds, leadership changes, incidents, new capabilities and announcements across the vendors in this catalog — collected automatically from vendors' own public feeds. Product release versions are on their own page, so the landing feed here never reads as one vendor's changelog.Where this comes fromEvery item links to the vendor's own published source: a GitHub release feed, a changelog, a status page, an SEC filing or a company blog. Nothing here is written or editorially reviewed by 0-Doubt, and nothing is a recommendation — it is machine-collected public record, labelled the same way as all other baseline content on the platform. Coverage is uneven by design: a vendor that publishes no public feed will not appear, and that absence is not a judgement about the vendor.
25,301 matching stories — showing up to 3 per vendorPer-vendor capWhile browsing, no single vendor may fill more than three slots on any one page, so a vendor with a chatty release cadence cannot buy the page you are looking at. That means a page can render fewer stories than the count above. Searching a headline or naming a vendor turns the cap off, because then you have asked to see everything that matches.
Twenty Years Later: Perspective on the 2006 Liquid Bomb Plot
Twenty years ago, British authorities uncovered a terror plot to smuggle liquid explosives onto transatlantic flights and take down multiple passenger planes bound for the United States. A recent UK documentary highlighted this story and includes an interview with our Co-Founder and Executive Chairman Michael Chertoff. Published by ITV News, the show details how intelligence […] The post Twenty Years Later: Perspective on the 2006 Liquid Bomb Plot appeared first on The Chertoff Group.
N-central Security Update – August 10, 2026
Attackers used to work at human speed. A skilled researcher found a vulnerability, built an exploit, and chained it into an attack over days or weeks. That gave defenders room to scan, triage, and patch. That room is gone. AI hasn't just changed how IT teams work, it's changed how attackers operate. Vulnerability discovery, exploit weaponization, and attack chaining that once took weeks now happen in hours. Unfortunately, the monthly patch cycles and manual triage that most teams still run were
BlackFile Explained
How vishing and SSO compromise lead to mass data theft.
CMMC phase 2 is paused. The email problem isn’t.
CMMC Phase 2 is paused, but CUI still moves through email, chat, and collaboration tools. Here’s what contractors should focus on now.
Ransomware & Remote Access
Why the decisive moment is an access decision, not a ransom note.
CrowdRecon is coming: turning hacker reconnaissance into security intelligence
At DEF CON 34, our team introduced something exciting. Something the Intigriti team has been building for months, and our Senior Product Manager, Radu Voloaga, took to the stage in the Bug Bounty Village to give everyone the first real look at CrowdRecon. How CrowdRecon closes the gap It started with a question we kept running into: what happens to all the reconnaissance work hackers do before a vulnerability is even found? That work rarely makes it into a report, but it's often where the real s
Beyond CVSS: Contextual Vulnerability Intelligence for CVEs
CVSS rates worst-case severity, not real risk. ZeroPath layers four signals on every dependency CVE: the real advisory CVSS, CISA KEV, FIRST EPSS, and an AI-derived contextual severity for your repo, refreshed daily and kept honest.
Failures or Failsafe?: A Perspective from the Arena of AI Detection Evals
DeepTempo’s cybersecurity foundation model, LogLM, played an integral role in discovering errors in a variety of datasets it was being evaluated on. Beyond proving superior performance and exposing critical flaws in public data, these rigorous evaluations and data quality checks power continuous model improvements.
AI Runtime Protection: Governing What Agents Do, Not Just What They're Told
Island AI Runtime Protection gives enterprises the visibility and control to govern every action an agent takes
Accelerating P&C Delivery with Developer Assistants from Guidewire
Lessons from the July Water Utility Attacks: What the Joint Advisory Tells Operators to Do Next
Secure communications help water utilities act fast during cyber incidents, coordinating response, containment, and continuity when trusted channels matter most.
The readiness gap: Banking's AI ambition outruns persistent capability
Executives in every industry are pushed to move faster on AI, but fewer are asked to move more cautiously or safely. The risks in banking feel higher because the regulators are often right at the door and decisions involve other people's money and livelihoods. Of course, the challenges in financial services are also facing healthcare, insurance and many other industries. Research shows the ambition-vs-readiness gap is increasing pressure in banking. A global survey of nearly 1,500 senior…
Workload identity vs secrets: When to use each
Workload identity and secrets managers solve different problems. Here's which governs which credential, and how to run them together.
Deception and the OpenAI / Hugging Face agentic breach | Tracebit
At Black Hat, OpenAI named honey tokens and deception as a key defensive investment against offensive AI agents. We map the publicly disclosed TTPs from the OpenAI / Hugging Face incident against Tracebit deception, stage by stage.
Browser threat landscape: mid-year update 2026
PhaaS industrialization, Scattered Spider copycats, and AI-augmented tooling — what the threat landscape looks like in 2026 so far.
Why energy companies need a unified namespace now
Energy operators don't lack data, they lack trust in it. Discover why a unified namespace fixes fragmented KPIs, and where it genuinely isn't needed yet.
Instrument serverless apps with agentic onboarding
Use Datadog agentic onboarding to instrument AWS Lambda, Google Cloud Run, and Azure Container Apps from an AI assistant or CLI.
How to manage risk from unfixed Kubernetes CVEs
Learn how to confirm whether your cluster is exposed to unfixed Kubernetes CVEs and build detection queries using Kubernetes audit logs.
What is device fingerprinting? The definitive 2026 guide
Device fingerprinting identifies a device from 250+ browser, hardware and network signals that survive cookie-clearing, incognito mode and VPN use.
How to detect bots on your website in 2026
Detecting bots in 2026 means combining network, browser, and behavioral signals into one verdict that also flags AI agents like OpenAI Operator.
TXTBOOK A Supply Chain Heist, Rehearsed in Public
A Dependency Confusion Campaign Against Tinkoff Bank's Private Package Namespace Authors :
security.txt in Deutschland: 1,8 Prozent der Webseiten haben einen Meldeweg
Prüfen Sie, ob Ihre Domäne eine security.txt bereitstellt: so erfüllen Sie die CRA-Meldepflicht und erleichtern Sicherheitsforschern den Meldeweg.
Governance Is What Lets Enterprises Move Faster on AI
Governing agentic AI takes visibility, control, and accountability. See how continuous AI governance and runtime enforcement work once agents reach production.
How Two Small Bugs Led to a Critical Vulnerability and a Cryptography Audit of Go’s SSH Library
Learn about NCC Group's cryptographic audit of Go's SSH package.
AI Protect: Governing What Agents Do, Not Just What They're Told
Island AI Protect gives enterprises the visibility and control to govern every action an agent takes
Splunk AI: How an AI SOC Agent Cracked a Dead-End Alert in 2m 38s
Splunk AI meets an autonomous AI SOC agent: it worked a Splunk Enterprise Security alert with no EDR, found a 50-host campaign, and hit a 91% verdict in 2m 38s.
Show, Don't Tell: What Evo Continuous Offensive Security Found in a Real Enterprise SaaS
A real Evo Continuous Offensive Security assessment uncovered 33 confirmed vulnerabilities in a multi-tenant enterprise SaaS, including tenant-wide compromise and critical authorization flaws.
JupiterOne Joins Anthropic's MCP Directory | JupiterOne
JupiterOne's MCP server is now listed as a community server in Anthropic's MCP directory. Here's what it enables and how to connect Claude to your security graph.
Sophos Working with OpenAI on security from AI, with AI, and for AI
Sophos to bring OpenAI’s frontier models to the channel and managed service providers in newly announced partnership
Security at the Speed of Agentic AI
Agents act across gateways, tools, and paths that traditional controls don't fully see. Learn how to secure the full agent workflow, so teams can adopt and build AI agents with risk under control. The post Security at the Speed of Agentic AI appeared first on Aurascape, Inc..
Agentless DLP: What It Means and What Each Approach Covers
“Agentless DLP” has become one of the most overloaded terms in enterprise security. At least five architecturally different approaches use the label, and they cover very different exfiltration paths. If you’re evaluating agentless DLP because you have contractors, BYOD users, or GenAI adoption you can’t put an endpoint agent on, the type matters more than […] The post Agentless DLP: What It Means and What Each Approach Covers appeared first on RedAccess.
Shai-Hulud Outbreak Debrief: The Worm Evolves into MCP
5 days after 440+ npm packages were compromised, 5 malicious repositories remain live in the wild—and the threat is still active While the cybersecurity community was traveling to Black Hat last week, threat actors unleashed one of the most widespread software supply chain attacks of 2026. A new evolution of the Shai-Hulud self-propagating worm compromised […] The post Shai-Hulud Outbreak Debrief: The Worm Evolves into MCP appeared first on OX Security.
Internal Audit Applicability in India: Rules, Requirements & Thresholds
Short Answer: Internal audit is mandatory in India for prescribed classes of companies under Section 138 of the Companies Act, 2013 and Rule 13 of the Companies (Accounts) Rules, 2014. Every listed company is covered, while specified unlisted public and private companies are covered when they meet the applicable turnover, paid-up capital, borrowing or deposit... The post Internal Audit Applicability in India: Rules, Requirements & Thresholds appeared first on ASPIA Infotech.
Email Authentication Methods: How SPF, DKIM, and DMARC Protect Your Domain
Learn what email authentication is and how to authenticate your emails to prevent spoofing, boost deliverability, and protect your brand.
GhostJacking Attacks: Half of the Fortune 500 Run These Tools. Getting Blocked by the Firewall Was the Way to Take Over Their AI Agents
Half the Fortune 500 run the tools that let us in. It will be presented at DEFCON, the largest hacker conference, where we’ll show how a request their own firewall blocked was the way in.
BLOCKCHAIN C2 AND FILELESS EXECUTION MAKE MAGECART HARDER TO DISRUPT
Blockchain C2 and Fileless Execution Make Magecart Harder to Disrupt A compromised checkout page can now take instructions from the blockchain, allowing Magecart operators to change malicious infrastructure without touching the infected website again. In this campaign, attackers used a Polygon smart contract to dynamically resolve the domain hosting the next-stage skimmer, making the operation The post BLOCKCHAIN C2 AND FILELESS EXECUTION MAKE MAGECART HARDER TO DISRUPT appeared first on Source
Emerging Threat: (CVE-2026-58048) cPanel & WHM Database Privilege Escalation via Database Rename
A privilege escalation flaw in cPanel & WHM's database management lets any authenticated hosting account execute database commands with full administrative privileges, exposing every database on the server.
Tampa Electric Selects Ericsson, OneLayer, and Burns & McDonnell to Deploy, Manage, and Secure Its Private LTE Network
Deployment brings SIM lifecycle management, device visibility, and Zero Trust security to Tampa Electric’s private LTE infrastructure powered by Ericsson TAMPA, Fla. and KANSAS CITY, Mo. and BOSTON, Aug. 06, 2026 (GLOBE NEWSWIRE) — Tampa Electric Company (TECO), one of the first investor-owned utilities to deploy a private LTE network for operational use, selected Ericsson and OneLayer to deploy, manage and […] The post Tampa Electric Selects Ericsson, OneLayer, and Burns &
Is the AI Security Stack Missing the Interaction Layer?
Network controls have gotten sharper about AI in the last year. Application control that identifies GenAI apps by protocol signature. Telemetry parsing for HTTP-based Model Context Protocol streams. Some implementations now add OCR to inspect sensitive data hidden inside image uploads. Real advances, all of them. But only when the traffic actually reaches an enforcement […] The post Is the AI Security Stack Missing the Interaction Layer? appeared first on RedAccess.