October 13, 2026

What happened

The article lists a CISO & Head of IT role at Trigo in Israel, with responsibility for leading IT and information security, setting strategy and priorities, managing budgets, and reporting progress to management. [1]

The Trigo role covers corporate infrastructure, IT support, identity and access management, security operations, and incident response. [2]

The article also lists a Cloud Threat Hunter role at Synechron in the United States with a hybrid work arrangement. [3]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   CISO & Head of IT Trigo | Israel | On-site – View job details As a CISO & Head of IT, you will lead the IT and Information Security team, set strategy and priorities, manage budgets and report progress to management.

2. [2]

   You will oversee corporate infrastructure, IT support, identity and access management, security operations and incident response.

3. [3]

   Cloud Threat Hunter Synechron | USA | Hybrid – View job details As a Cloud Threat Hunter, you … More → The post Cybersecurity jobs available right now: October 13, 2026 appeared first on Help Net Security .

Read the original article →

What happened

Sola Security researchers Leon Goldberg and Gal Engelberg built an AI agent to re-grade findings raised by third-party cloud security platforms. [1] [2]

The agent was run on 9,967 HIGH findings from two commercial platforms in eight production environments; the supplied text is truncated after this description. [3]

Why it matters

The research examined whether checking each finding against the environment where it resides changes the severity ordering assigned by cloud security platforms. [1] [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Research from Sola Security tested what happens to that order when each finding gets checked against the environment where it lives.

2. [2]

   The researchers, Leon Goldberg and Gal Engelberg, built an AI agent that re-grades findings raised by third-party cloud security platforms.

3. [3]

   They ran it on 9,967 HIGH findings from two commercial platforms in eight production … More → The post Most high-severity cloud security findings are downgraded once an AI agent checks the environment appeared first on Help Net Security .

4. [4]

   Cloud security platforms turn every failed rule into a finding, and those findings pile up in queues sorted by severity.

Read the original article →

What happened

Booz Allen tested frontier AI models against an isolated industrial-control testbed in its Chantilly, Virginia lab and reported that they completed all eight attack objectives defined for the test. [1]

The reported activities included network mapping, associating equipment with known weaknesses, writing working exploits, pivoting into production-control systems, and taking administrative control of and moving a robotic arm. [2]

Why it matters

The test demonstrated successful model actions across multiple stages of an OT attack scenario, including exploitation, lateral movement into production systems, and physical manipulation of equipment. [1] [2]

Known limitations

The supplied evidence does not state which models, attack objectives, specific vulnerabilities, or test conditions beyond the isolated testbed were used. [1] [2]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Booz Allen set frontier AI models against an isolated industrial-control testbed in its Chantilly, Virginia lab and watched them clear all eight attack objectives it had defined.

2. [2]

   The Booz Allen frontier AI OT test architecture (Source: Booz Allen) The models mapped the network, tied equipment to known weaknesses, wrote working exploits, pivoted into the systems that run production, and in the last test found a robotic arm, took administrative control, and moved it.

Read the original article →

What happened

VMware recently moved to monthly software updates, partly to address bugs found using AI, while users had struggled to keep up because testing patch impacts takes time. [1] [2] [3]

HyperArmor is a planned add-on to VMware Cloud Foundation (VCF) that provides virtual patching, enforcement of known-good VCF states, and multifactor authentication. [4] [5] [6] [7] [8]

VMware says HyperArmor will block malicious traffic directed at VCF implementations without changing vulnerable software, similar to virtual patching in vDefend. [5] [9] [10] [11]

HyperArmor is designed to protect against attackers using stolen privileged credentials to access the ESX hypervisor by requiring SSH credentials and a secondary time-based one-time password sent to a verified user. [12] [13]

When a human approves a change to a known-good VCF state, HyperArmor updates its policies to allow that altered state. [6] [14]

Why it matters

The article reports that HyperArmor was developed by Broadcom security businesses Symantec and Carbon Black, although VMware will sell it as a VCF add-on. [8] [15] [16]

Broadcom's AgentMinder, launched for VCF and third-party Kubernetes stacks, aims to stop AI running inside VCF from going rogue. [17] [18]

Known limitations

The supplied evidence does not state HyperArmor's availability date, pricing, or whether it has been independently tested against attacks. [8]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   VMware has discovered so many flaws in its products using AI that it felt the need to create a new product to help users manage them.

2. [2]

   Prashanth Shenoy, vice president of product marketing in the VMware Cloud Foundation (VCF) division of Broadcom, told The Register that VMware recently moved to deliver monthly software updates, in part to address bugs found by AI.

3. [3]

   He admitted, however, that users struggled to keep up with the pace of change before the move to monthly releases because it takes time to test the impact of patches – even though VCF allows non-disruptive live patching.

4. [4]

   Shenoy said customers therefore need "room to breathe." Enter a new tool called HyperArmor, which does three things.

5. [5]

   One is "virtual patching" – blocking attempts to exploit a vulnerability without changing the vulnerable software itself.

6. [6]

   Another job HyperArmor takes on is rejecting any actions that alter known good states in VCF.

7. [7]

   The third job is multi-factor authentication (MFA).

8. [8]

   HyperArmor is a discrete product that VMware will sell as an add-on to VCF.

9. [9]

   VMware already offers virtual patching in its vDefend network security product and explains how it works by referring to CVE-2024-5806, the bug in file transfer tool MOVEIt that made waves in 2024.

10. [10]

    VMware says vDefend can detect patterns associated with attempts to exploit the authentication bypass flaw and block the traffic.

11. [11]

    HyperArmor will do similar things to block malicious traffic heading toward VCF implementations.

12. [12]

    He agreed but explained that VMware designed HyperArmor to protect against attackers using stolen privileged credentials to access VMware's ESX hypervisor.

13. [13]

    To (hopefully) stop attackers using those creds to take over a host, HyperArmor requires users to input SSH credentials and a secondary time-based one-time password sent to a user with a verified identity.

14. [14]

    If a human approves a change, the tool will update policies to allow the altered state.

15. [15]

    While VMware will sell HyperArmor, it didn't create the product.

16. [16]

    Broadcom's security businesses Symantec and Carbon Black cooked it up.

17. [17]

    At VMware Explore on August 31, Broadcom launched AgentMinder, which aims to stop AI running inside VCF going rogue.

18. [18]

    AgentMinder came from CA but was launched as a tool for VCF (and third-party Kubernetes stacks).

Read the original article →

What happened

The article describes a scam text containing only the recipient’s name followed by a question mark, with no link, payment demand, account warning, or fake image. [1] [2] [3] [4] [5] [6]

The sender’s possession of a correct name and phone number does not establish familiarity; the article says such combinations can come from breached data traded by cybercriminals. [12] [13]

The article says a wrong-number-style conversation can develop into relationship-building and a fraudulent investment solicitation, and attributes similar cryptocurrency investment-scam approaches using unsolicited texts to the FBI. [14] [15] [16] [17] [18] [19]

Why it matters

The article says scammers use brief messages to prompt engagement, which can lead to requests for money or personal information; a reply may also show that the number is responsive and be reused or sold for other campaigns. [7] [8] [9] [10] [11]

Source-supported guidance

The article recommends reviewing messaging-app and social-media privacy settings and limiting visibility of profile details, status, last-seen information, and phone number. [20] [21] [22]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   The one we got this week was a lot simpler.

2. [2]

   No suspicious link.

3. [3]

   No overdue payment.

4. [4]

   No warning that your account is about to be closed.

5. [5]

   No fake images.

6. [6]

   Just your name, followed by a question mark.

7. [7]

   A message like this invites an almost automatic response: “Yes?” Or perhaps, “Who’s this?” That’s because a scammer needs to get you to engage before they can persuade you to hand over money or personal information .

8. [8]

   Sometimes, the first thing they ask you to give them is simply a reply.

9. [9]

   Even if the conversation goes no further, your reply can still be useful to the scammer.

10. [10]

    As we explained in our earlier article on wrong-number texts , it shows that you are willing to engage.

11. [11]

    Your number can then be reused or sold for other scam campaigns.

12. [12]

    It’s tempting to think, “They have my name and number, so this must be someone I know.” But having those details is not evidence they know you.

13. [13]

    As we discussed in the Marketplace article, name-and-phone-number combinations can come from breached data traded by cybercriminals.

14. [14]

    What follows depends on the scheme.

15. [15]

    The first message doesn’t tell us which route this scammer planned to take.

16. [16]

    But documented scams show how an innocuous introduction can develop: A supposed wrong-number exchange becomes a friendly conversation.

17. [17]

    The sender builds a relationship rather than immediately asking for money.

18. [18]

    An investment opportunity eventually enters the discussion.

19. [19]

    The victim is encouraged to send money to a fraudulent investment scheme.

20. [20]

    Make yourself harder to profile.

21. [21]

    Review privacy settings on messaging apps and social media.

22. [22]

    Limit who can see your profile photo, status, last seen, and phone number.

Read the original article →

What happened

GitLab released patches for a maximum-severity flaw affecting its Community Edition and Enterprise Edition software that could enable arbitrary code execution. [1]

Under certain conditions, an unauthenticated user could execute arbitrary code on a GitLab server by supplying a specially crafted regular expression in a CI/CD context. [2]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   GitLab has released patches for a maximum-severity security flaw impacting its Community Edition (CE) and Enterprise Edition (EE) software that could result in arbitrary code execution.

2. [2]

   "GitLab has remediated an issue that, under certain conditions, could have allowed an unauthenticated user to execute arbitrary code on the GitLab server via a specially crafted regular expression in a CI/CD

Read the original article →

What happened

Two of 31 newly disclosed flaws in Progress Software’s MarkLogic Server could allow an account with some management privileges to take administrator control of the server. [1]

Progress published all 31 flaws on October 12 and rated five of them critical. [2]

Fixes have been available since September 10 in MarkLogic 11.3.7 and MarkLogic 12.1.0. [3]

Known limitations

The supplied evidence does not identify the specific flaw identifiers, affected configurations, or whether exploitation was observed. [1] [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Two of 31 newly disclosed flaws in Progress Software's MarkLogic Server database could let an account given only some management privileges take administrator control of the server, the company says.

2. [2]

   Progress published all 31 on October 12 and scored five of them critical.

3. [3]

   Fixes have been available since September 10, in MarkLogic 11.3.7 and MarkLogic 12.1.0.

4. [4]

   Affected

Read the original article →

What happened

Attackers reportedly took remote control of refrigeration equipment at hundreds of French supermarkets on October 11 and raised temperatures, causing stores to discard food. [1]

The attackers reached the fridges and freezers through a controller that manages all refrigeration units in a store; maintenance firms can manage the Danfoss-made device remotely. [2]

Why it matters

The incident affected supermarket refrigeration operations and resulted in food loss, according to several retail chains. [1]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Attackers took remote control of fridges and freezers at hundreds of French supermarkets on the night of October 11 and warmed them, forcing stores to throw food away, several retail chains said.

2. [2]

   The attackers reached them through the controller that runs all of a store's fridges and freezers, a device made by Danish company Danfoss that maintenance firms can manage remotely.

Read the original article →

What happened

Bloom Security researchers found that npm Trusted Publishing allowed an attacker to push new versions of another party’s package without stealing a credential, by registering a GitHub repository name the maintainer had left behind. [1] [2]

GitHub, which owns npm, said it had shipped fixes for the issue. [3]

Why it matters

Bloom counted more than 1,500 npm packages exposed to this takeover path, with more than 10 million total downloads. [3] [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Bloom Security researchers found that npm’s Trusted Publishing feature let an attacker push new versions of someone else’s package without stealing a credential.

2. [2]

   The attacker only needed to register a GitHub repository name the maintainer had left behind.

3. [3]

   Bloom counted more than 1,500 npm packages exposed this way, and GitHub, which owns npm, says it has shipped fixes.

4. [4]

   The exposed packages have more than 10 million downloads in total and more than a million a … More → The post npm Trusted Publishing flaw exposed 1,500+ npm packages to takeover appeared first on Help Net Security .

Read the original article →

What happened

Meta’s Muse Wonder and Muse Wonder Full are AI-agent models intended for tasks including preparing insurance claims, reviewing leases, and developing software. [1]

The models combine document analysis and tool use with retaining relevant details across conversations and working through obstacles toward a user’s goal. [2]

They are intended for commercial and research use and support writing, reasoning, coding, planning, and research. [3]

The models can process text, images, video, and audio, and produce [truncated]. [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Meta’s Muse Wonder and Muse Wonder Full models are built for AI agents that handle tasks such as preparing insurance claims, reviewing leases and developing software.

2. [2]

   The models combine document analysis and tool use with the ability to retain relevant details across conversations and work through obstacles toward a user’s goal.

3. [3]

   Intended for commercial and research use, they support writing, reasoning, coding, planning and research.

4. [4]

   They can process text, images, video and audio, and produce … More → The post Meta outlines Muse Wonder safeguards for AI agents acting on your behalf appeared first on Help Net Security .

Read the original article →

What happened

The walkthrough describes AI use through a browser chat assistant, an editor extension, and a local model runtime. [1] [2] [3]

Why it matters

The examples include AI activity that may not appear in an installed-programs list, including browser-based chat and editor extensions. [1] [2]

A local model runtime runs models directly on the machine rather than sending data to an external service, and is presented as an option when company data should not be sent externally. [3]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   A user opens a browser tab, pastes data into a chat assistant, and nothing is installed.

2. [2]

   A developer adds a coding assistant as an editor extension, and nothing shows up in the installed programs list.

3. [3]

   Someone downloads a local model runtime, a program that runs models directly on the machine instead of sending data to an external service, because they were told not to send company data

Read the original article →

What happened

Three LG webOS vulnerabilities can be chained by an attacker on the same local network to obtain root access to certain smart TVs without a PIN or owner interaction. [1]

The chain was tested on the LG 32LQ636BPSA and 27TQ600SW models running webOS 7.5.7-25; the supplied text is truncated after “on firmware.” [2]

LG has issued a fix, according to an advisory published by SSD Secure Disclosure on October 12. [3]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Three vulnerabilities in LG's webOS can be chained to give an attacker on the same local network root access to certain smart TVs without a PIN or any interaction from the owner.

2. [2]

   The chain was tested on two TV models, the 32LQ636BPSA and the 27TQ600SW, running webOS 7.5.7-25 on firmware

3. [3]

   LG has issued a fix, according to an advisory published by SSD Secure Disclosure on October 12.

Read the original article →

What happened

Mozilla’s paid Firefox Enterprise platform is designed to help organizations control sensitive-data sharing, assess device security, and manage employee access to web applications. [1]

The platform pairs the browser with a management dashboard hosted on the organization’s own infrastructure. [2]

IT teams can set workforce policies, restrict AI tools, and connect browser activity to existing security-monitoring systems. [3]

Why it matters

Mozilla is targeting organizations with demanding security and regulatory requirements. [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Mozilla’s Firefox Enterprise is a paid platform that lets organizations control how sensitive data is shared, check device security, and manage employee access to web applications.

2. [2]

   The platform pairs the browser with a management dashboard hosted on an organization’s own infrastructure.

3. [3]

   IT teams can set policies across their workforce, restrict AI tools, and connect browser activity to existing security monitoring systems.

4. [4]

   Mozilla is targeting organizations with demanding security and regulatory requirements.

Read the original article →

What happened

Ridge Security announced RidgeOne, a self-contained solution for autonomous AI-powered penetration testing and agentic red teaming aimed at mid-sized organizations. [1]

RidgeOne combines Ridge Security’s RidgeGen agentic security validation platform with an open-weight AI model running on NVIDIA DGX Spark. [2]

Why it matters

The solution is described as enabling security testing within an organization’s own environment without sending sensitive testing data to an external AI provider or incurring ongoing AI token costs. [2]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Ridge Security has announced RidgeOne, a self-contained solution that brings autonomous AI-powered penetration testing and agentic red teaming to mid-sized organizations.

2. [2]

   Combining Ridge Security’s RidgeGen agentic security validation platform with an open-weight AI model running on NVIDIA DGX Spark, RidgeOne enables organizations to perform deep security testing within their own environment without sending sensitive testing data to an external AI provider or incurring ongoing AI token costs.

Read the original article →

What happened

Lattice Semiconductor released version 5.0 of its Lattice Sentry solution stack, adding post-quantum cryptography to platform firmware resilience and attestation capabilities. [1]

The update is described as providing CNSA 2.0-aligned, NIST-compliant protection for AI data centers and critical infrastructure, with a validated, production-ready approach to quantum-resistant security. [2]

Why it matters

The source characterizes platform firmware as a primary attack vector and says the quantum era will significantly change established security approaches. [3] [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Lattice Semiconductor has released version 5.0 of its Lattice Sentry solution stack, adding post-quantum cryptography (PQC) to its platform firmware resilience (PFR) and attestation capabilities.

2. [2]

   The update provides CNSA 2.0-aligned, NIST-compliant protection for AI data centers and critical infrastructure, giving system designers a validated, production-ready approach to quantum-resistant security.

3. [3]

   “AI infrastructure and the threat landscape are evolving rapidly.

4. [4]

   Platform firmware has become a primary attack vector, and the quantum era will upend decades of security … More → The post Lattice Sentry 5.0 brings quantum-resistant firmware security to AI data centers appeared first on Help Net Security .

Read the original article →

What happened

NinjaOne announced Katana AI, described as an agentic AI technician supporting IT workers, alongside an MCP server that extends the platform’s governed access to AI tools used by teams. [1]

Why it matters

IDC research VP Phil Hochmuth said modern IT environments generate more data than any technician can process. [2]

The announcement is presented as intended to accelerate IT problem resolution by reducing time spent searching for problem context rather than solving issues. [1] [3]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   NinjaOne has announced Katana AI, an agentic AI technician that supports and elevates IT workers, and the NinjaOne MCP server, an agentic orchestration connector that extends the platform’s governed access to the AI tools teams already use.

2. [2]

   “Modern IT environments generate more data than any technician can process,” said Phil Hochmuth, research VP, endpoint management and enterprise mobility at IDC.

3. [3]

   “Hunting for a problem’s context instead of actually solving the issue wastes time for IT … More → The post NinjaOne combines Katana AI and MCP server to accelerate IT problem resolution appeared first on Help Net Security .

Read the original article →

What happened

Trellix announced Trellix Wise for Private Cloud, described as an on-premises, federated AI core for agentic security operations. [1]

The solution extends a security fabric across existing organizational infrastructure to improve threat detection and response while preserving data sovereignty, governance, and predictable costs. [2]

Why it matters

Trellix CEO Vishal Rao characterized AI as accelerating and commercializing attacks, framing the offering within a changed threat landscape. [3]

Known limitations

The supplied evidence does not provide technical deployment details, evaluation results, availability conditions, or specific security-team workflows. [1] [2] [4]

Cited source evidence

These excerpts support the numbered claims above. They come from one source article and do not establish independent corroboration.

1. [1]

   Trellix has announced Trellix Wise for Private Cloud, an on-premises, federated AI core for agentic security operations.

2. [2]

   The solution extends a security fabric across an organization’s existing infrastructure to improve threat detection and response while preserving data sovereignty, governance, and predictable costs.

3. [3]

   “The threat landscape has fundamentally shifted, where AI isn’t just accelerating attacks; it’s commercializing them,” said Vishal Rao, Chief Executive Officer, Trellix.

4. [4]

   “Our strategy for the AI era is clear: meet AI-powered threats … More → The post Trellix Wise for Private Cloud delivers auditable AI investigations for security teams appeared first on Help Net Security .

Read the original article →