Live

Intelligence Feed

Latest threat intelligence articles from trusted security sources, auto-processed to extract entities, IoCs, and TTPs.

Filtered by source: bleeping-computer Clear filter
TSMC employee data breach linked to cloud provider

2d ago · bleeping-computer

Taiwan Semiconductor Manufacturing Company (TSMC) confirmed a data breach involving personal information of current and former employees, linked to a compromise of a third-party cloud service provider. The stolen data, totaling 6.9GB and reportedly containing 46,000 employee records with sensitive details such as national ID numbers, addresses, contact information, and banking data, was put up for auction by a threat actor using the alias 'SmithBianl'. The actor posted a sample of 499 records on a hacker forum and demanded 3 BTC (~$65,000) for the full dataset, threatening to release it publicly if no buyer emerged. TSMC stated there is no evidence its own systems were compromised, and the breach originated outside its infrastructure.

1 IoCs
FBI warns ARTEX AI hacking tool is powering global cyberattacks

2d ago · bleeping-computer

The FBI has issued a warning about the misuse of ARTEX, an AI-powered open-source penetration testing framework, by multiple threat actors to automate cyberattacks globally. The tool enables rapid reconnaissance, exploitation, privilege escalation, and data theft with minimal human intervention, significantly accelerating attack timelines—some intrusions reach privileged access in under three hours. ARTEX has been used in high-profile incidents including a ConnectWise supply-chain breach affecting Microsoft 365 tenants, attacks on South Korean financial institutions, and a cyberespionage campaign by group Vapor Panda targeting over 200 organizations. Although the original GitHub repository was taken down, the framework remains accessible via forks and a commercial version called ARTEX PRO.

1 IoCs 1 Actors
US disrupts North Korean campaign behind $400 million crypto theft

3d ago · bleeping-computer

The U.S. government disrupted a North Korean social engineering campaign named 'FaceStealer', attributed to the BlueNoroff group, which stole over $400 million in cryptocurrency. The campaign used deepfakes, fake technical fixes, and compromised accounts on Telegram, LinkedIn, Slack, and Discord to lure victims into video meetings where their webcams were secretly recorded. These recordings were later reused to manipulate new targets into downloading malware via malicious links, leading to crypto theft and credential harvesting. The operation targeted individuals in the cryptocurrency, Web3, and open-source development communities across 58 countries, with infections sometimes delayed for over a year to maximize financial gain.

1 Actors
Cursor confirms security breach led to malicious Claude Code extension

3d ago · bleeping-computer

Cursor confirmed a security breach stemming from compromised credentials at AI infrastructure provider Fireworks AI, which allowed attackers to distribute malicious versions of the Claude Code extension through Cursor's marketplace. The malicious extension, versions 2.1.312 and 2.1.313, was designed to install persistent SSH backdoors on developers' systems by deploying attacker-controlled SSH keys and executing malicious JavaScript. Up to 5,000 users were potentially exposed during the 11-hour window the malicious extensions were live. The breach also led to the theft of customer data, including unredacted API credentials from AI coding-agent traces, which were used to compromise additional systems.

2 IoCs
Check Point warns of zero-day flaw exploited for root access

4d ago · bleeping-computer

Check Point has disclosed and patched a zero-day vulnerability, CVE-2026-123039, that has been actively exploited since late October 2026 to achieve local privilege escalation to root on vulnerable devices. The flaw exists in the Gaia OS DynamicCLI daemon, which listens on a world-writable socket without authentication, allowing authenticated local users to execute commands as root. Attackers initially compromised systems via the Mobile Access portal using PHP2Shell vulnerabilities, then used the zero-day to escalate privileges, deploy webshells, and install TSGA backdoor and TSProxy malware before deploying wiper malware on VMware ESXi hosts.

1 IoCs
Check Point gateways hacked via PHP2Shell to deploy VMware ESXi wiper

4d ago · bleeping-computer

Vulnerable Check Point Security Gateway instances were exploited in attacks leveraging the PHP2Shell vulnerabilities (CVE-2026-109618, CVE-2026-109619) to gain unauthenticated remote code execution. Attackers deployed PHP webshells and the TSGA backdoor with TSProxy to establish persistent access, then moved laterally to VMware ESXi hosts using the gogo scanner. A data-wiping malware named destroy_v6 was deployed on ESXi hosts, which performed recursive deletion and zero-filling of data to hinder recovery. A political message in DREAM_ANNEXATION.txt rejected ransom negotiations, indicating destructive intent. The second wave of attacks did not include destructive actions, and attribution remains uncertain despite tentative links to the PhiliKit framework and UNC5221.

3 IoCs 1 Actors 2 CVEs
German military drone operations disrupted by cyberattack

4d ago · bleeping-computer

German defense technology company Quantum Systems suffered a cyberattack in late October 2026, attributed to hostile actors targeting its software update infrastructure. The attackers compromised the company's QBase ground-control and mission-planning software by delivering a malicious update through the vendor's update distribution server. This led to the grounding of Quantum Systems drones by the German armed forces as a precaution, disrupting military drone operations. The incident is under investigation by German military counterintelligence (MAD), the domestic intelligence service (BfV), and the federal cybersecurity agency (BSI).

1 IoCs
Trezor confirms crypto-stealing implants in three wallets, releases fix

3d ago · bleeping-computer

Trezor has confirmed that three Safe 5 hardware wallets sold by Southeast Asian reseller CryptoBilis contained malicious hardware implants designed to steal wallet recovery phrases and transmit them over cellular networks. The implants were located on an additional circuit board inserted between the main board and display, enabling them to capture sensitive data such as recovery words and passphrases without modifying legitimate firmware. Trezor released firmware version 2.13.1 on November 11, 2026, which blocks the functionality of these implants, though it cannot protect against data already exfiltrated. No fund losses have been reported, but affected customers are advised to migrate to new wallets with fresh recovery phrases.

Apify rotates API tokens, prompts 2FA resets after breach

5d ago · bleeping-computer

Apify detected unauthorized access to its internal systems between November 9 and 10, 2026, leading to the rotation of internal secrets, API tokens, and customer 2FA resets as a precaution. The company revoked OAuth connections to third-party services, including GitHub, to mitigate potential downstream risks. While Apify stated there was no evidence of customer account data being compromised, customer API tokens and Apify Proxy passwords were rotated, and users were advised to reset their 2FA due to potential exposure.

Police dismantle Gootloader, CastleRAT, StealC malware infrastructure

5d ago · bleeping-computer

Law enforcement agencies from nine countries, coordinated by Europol and Eurojust, dismantled infrastructure associated with the Gootloader, CastleRAT, and StealC malware operations as part of Operation Endgame. The operation resulted in the seizure of 257 domains and 235 servers used by these cybercriminal networks. Authorities identified over 6.4 million stolen victim datasets, including more than 6 million credentials, which were exfiltrated by the malware and intended for use in follow-on attacks such as ransomware. Victims' data will be shared with cybersecurity organizations to enable alerts and remediation, particularly urging WordPress users to update credentials and enable multi-factor authentication.

2 IoCs 1 Actors 12 Malware
New Windows SignDrop zero-day gives hackers SYSTEM privileges

5d ago · bleeping-computer

Security researcher Abdelhamid Naceri disclosed a Windows zero-day vulnerability named SignDrop that enables local privilege escalation to SYSTEM level by exploiting improper logic in the Common Log File System (CLFS) driver's authentication file handling. The flaw abuses Microsoft's own mitigation introduced in October 2025 to defend against CLFS-related authentication attacks, allowing attackers to trigger arbitrary file creation in the context of privileged processes such as Windows Defender. This can be used to bypass antivirus and EDR self-defense mechanisms. The vulnerability affects Windows 11 26H2, Windows 11 26H2 Experimental, and Windows Server 2025 Experimental, with no official patch available yet.

3 IoCs
Apache HTTP Server patches critical HTTP/2 remote code execution flaw

5d ago · bleeping-computer

A critical remote code execution (RCE) vulnerability has been patched in Apache HTTP Server's mod_http2 module, tracked as CVE-2026-119618. The flaw is caused by a race condition in the assign_headers() function that leads to heap memory corruption when HTTP/2 is enabled. Attackers can exploit it without authentication to execute arbitrary code with the web server's privileges on affected versions 2.4.55 through 2.4.70. Although no active exploitation has been observed yet, 64% of cloud environments contain vulnerable resources, and rapid patching is strongly advised.

New N-able "4EverAble2Shell" zero-day exploited hours after PoC released

4d ago · bleeping-computer

Hackers are actively exploiting a newly disclosed zero-day vulnerability in N-able N-central, dubbed '4EverAble2Shell', to gain root access to servers and deploy cryptocurrency miners. The exploit chain combines CVE-2026-90436 in Apache Axis with a privilege escalation flaw in N-central, allowing unauthenticated attackers to execute commands as root. Security firms Previdian and Huntback confirmed exploitation in the wild, with at least 21 production servers compromised, including those running the latest security update. Attackers are downloading Monero miners via shell scripts from a known malicious IP, using stealthy techniques to avoid detection.

6 IoCs 1 Malware 1 CVEs
Hackers compromise 4,800 GitHub accounts to steal cloud credentials

4d ago · bleeping-computer

A cyberattack campaign dubbed 'wfx-loot' compromised 4,837 GitHub accounts to target 61,764 repositories, aiming to steal CI/CD secrets and cloud credentials. The attacker used malicious pull requests with the title 'docs: dependency audit notes' to inject credential-stealing scripts into GitHub Actions workflows or npm install hooks. The script harvested GitHub tokens, cloud credentials (AWS, Azure, GCP), developer tool secrets, and infrastructure credentials, then exfiltrated them to attacker-controlled endpoints before rapidly reverting changes to evade detection. At least 1,835 repositories had secrets successfully stolen, and malicious code was published to Packagist and Go module repositories.

7 IoCs
New worm-like 'SCKit' malware targets MSPs in ScreenConnect attacks

4d ago · bleeping-computer

A new worm-like malware toolkit named SCKit has been used in recent attacks against managed service providers (MSPs) leveraging compromised ScreenConnect environments. The malware attempts to exploit a patched vulnerability, CVE-2026-84869, in ScreenConnect to execute malicious code on technician workstations during remote support sessions. The initial access vector appears to be a flaw in the RMM+ third-party extension, specifically versions 1.0.31 and 1.0.32, which allows unauthenticated command execution due to a blank access key authentication bypass. Although the SCKit toolkit was deployed and designed to spread worm-like by hijacking technician sessions and stealing authentication cookies, researchers have not observed successful exploitation in the wild due to existing patches.

1 IoCs
CISA flags Cisco BroadWorks flaw as actively exploited in attacks

6d ago · bleeping-computer

The US Cybersecurity and Infrastructure Security Agency (CISA) has added CVE-2026-20320, a high-severity XML External Entity (XXE) vulnerability in Cisco BroadWorks, to its Known Exploited Vulnerabilities (KEV) catalog due to active exploitation. The flaw allows unauthenticated remote attackers to read sensitive configuration files by sending malicious XML messages to the Open Client Interface - Provisioning (OCI-P) service. Threat intelligence firm Malbear Labs reported that a Chinese-speaking threat actor has exploited this vulnerability using a jailbroken DeepSeek Flash AI model in attacks targeting a European telecommunications provider and 13 connected organizations. The attackers accessed private keys and SSL credentials, enabling potential impersonation of internal BroadWorks components.

N-able releases security hotfixes for zero-day vulnerabilities

6d ago · bleeping-computer

N-able released emergency hotfixes for multiple zero-day vulnerabilities in its N-central remote monitoring and management (RMM) platform after a security researcher publicly disclosed the flaws without prior notification. The vulnerabilities, including critical RCE, XXE, OGNL injection, and SSRF flaws, were actively exploited in the wild, with attackers scanning for vulnerable systems and exploiting unauthenticated endpoints. As a precaution, N-able took all hosted N-central environments offline globally until patches were applied, urging on-premises customers to do the same and upgrade immediately.

Hackers use AntiFox AI agents in AWS credential-theft attacks

6d ago · bleeping-computer

Threat actors are leveraging self-hosted instances of AntiFox, an open-source AI-powered penetration testing platform, on Alibaba Cloud infrastructure to conduct attacks targeting AWS environments. The attackers deploy AI agents to perform automated tasks such as vulnerability scanning, webshell deployment, and AWS credential harvesting using valid IAM identities and EC2 roles. Observed malicious activities include deploying ASPX webshells on IIS servers, executing OS commands, exfiltrating configuration files, and making authenticated API calls to AWS services, resulting in the theft of production secrets in some cases.

1 IoCs 1 CVEs
Maersk confirms data breach as WhiteArmor ransomware demands $5 million

6d ago · bleeping-computer

Maersk has confirmed a data breach affecting a legacy system from its acquisition of LF Logistics, with the ransomware group WhiteArmor claiming responsibility and demanding $5 million in bitcoin. The attackers claim to have exfiltrated approximately 100GB of data, including sensitive information such as 30,000 driver records, plaintext passwords, 33 SUPER_ADMIN accounts, signed delivery receipts, purchase orders, and mobile device logs containing phone numbers, emails, and IMEI numbers. Maersk states the incident is isolated to a local system with no current evidence of impact on other systems or customer operations, though the investigation is ongoing.

Notion says Fireworks breach exposed customer AI chat transcripts

5d ago · bleeping-computer

A breach at Fireworks AI, the AI infrastructure provider for Notion, exposed customer AI chat transcripts from Notion's Insights feature. The attacker accessed Fireworks' development environment via a third-party cloud provider on October 23, compromising metadata, billing data, model weights, and customer datasets. Notion confirmed that the exposed data included unattributable AI chat transcripts from less than 1% of its workspaces, which were used for product analytics. Fireworks had previously experienced another breach in early October involving exposed API keys and tokens.

SAP patches ICM2SHELL and PROXYGATE maximum severity flaws

1w ago · bleeping-computer

SAP released its November 2026 security updates, patching 35 vulnerabilities, including four rated at maximum severity (CVSS 10/10). Two critical vulnerabilities, CVE-2026-76964 and CVE-2026-89126, are collectively tracked as ICM2SHELL by Onapsis, enabling unauthenticated remote code execution on SAP's NetWeaver and Web Dispatcher components. Another critical flaw, CVE-2026-89103, known as PROXYGATE, allows attackers to bypass authentication and gain full control of SAP systems by impersonating internal servers. These vulnerabilities pose severe risks as they require no user interaction or credentials, exposing core business data and transaction integrity.

1 CVEs
Binance warns of iOS flaw that could leak crypto wallet secrets

1w ago · bleeping-computer

Binance has warned of a vulnerability in iOS versions 26.7.1, 27.0.1, and 27.2 that allows malicious apps to bypass sandbox restrictions and access sensitive data from other apps, including cryptocurrency seed phrases, private keys, and login credentials. The issue, referred to as 'worse_query', enables unauthorized access to app storage without requiring a jailbreak. A proof-of-concept tool named WQ Files has been released, demonstrating how attackers can exploit this flaw to browse, export, edit, and delete data from other applications. Although no confirmed crypto losses have occurred, users are urged to update to patched iOS versions and remove untrusted apps.

1 IoCs
Critical Xen Orchestra auth bypass flaw exploited to hijack VM servers

1w ago · bleeping-computer

A critical authentication bypass vulnerability in Xen Orchestra, tracked as VSA-2026-062, is being actively exploited to gain administrator privileges and hijack XCP-ng virtualization hosts. The flaw allows attackers to bypass authentication by sending a specially crafted cookie containing an empty JSON object, enabling them to retrieve reusable login tokens and achieve full system compromise. Once inside, attackers can take over managed VMs and execute arbitrary code as root. Vates has released patched versions and provided indicators of compromise to help detect ongoing attacks.

7 IoCs
Microsoft November 2026 Patch Tuesday fixes 1,235 flaws, Outlook zero-day

1w ago · bleeping-computer

Microsoft's November 2026 Patch Tuesday addresses 1,235 security vulnerabilities, including one actively exploited zero-day in Microsoft Outlook. The zero-day, CVE-2026-124194, is a spoofing vulnerability that allows unauthenticated attackers to send a specially crafted message file, which when opened, causes Outlook to connect to an attacker-controlled network location without warning. This connection can lead to the disclosure of NTLM authentication hashes, enabling potential impersonation of the user. The attack requires user interaction and network connectivity, and exploits a known class of NTLM credential disclosure flaws often used in phishing campaigns.

Hackers exploit trusted AWS access to breach multiple cloud environments

6d ago · bleeping-computer

A threat actor compromised IAM credentials associated with Qubole's AWS account and exploited cross-account trust relationships to access multiple customers' AWS environments between October 14 and 26, 2026. The attacker assumed unused IAM roles in victim accounts, enumerated and exfiltrated data from S3 buckets, and harvested secrets from AWS Secrets Manager, SSM Parameter Store, and Lambda environment variables. Using TruffleHog, the attacker validated stolen credentials, including AWS keys, Microsoft Entra ID secrets, and Okta credentials, in a rapid, opportunistic campaign affecting dozens of organizations with millions of files exfiltrated from at least one victim.

Japanese furniture giant Nitori says hackers stole 37.6M customer records

6d ago · bleeping-computer

Japanese furniture retailer Nitori disclosed a cyberattack that led to the unauthorized access of up to 37.62 million customer records. The breach was detected on November 6, 2026, but attackers likely had access since mid-October. The compromised data includes personal information such as names, email addresses, postal addresses, phone numbers, dates of birth, and order details, but not passwords or credit card information. Nitori has blocked the access, notified authorities, and warned customers about potential phishing and scam attempts leveraging the stolen data.

Railway breach leads to $740,000 in fraudulent AI charges

6d ago · bleeping-computer

In October 2026, cloud hosting provider Railway suffered a breach where attackers exploited two vulnerabilities in its build infrastructure to gain access to internal credentials and customer environment variables. The threat actors used AI-powered agents to probe and exploit Railway's systems over a 26-hour period, ultimately stealing AI provider credentials from up to 100,000 workspaces. The stolen credentials were used to generate approximately $740,000 in fraudulent AI usage, with attackers likely targeting AI model distillation or resale. The incident is linked to the 'Black October' campaign, which has targeted multiple AI development platforms recently.

Hackers exploit SSHole OpenSSH flaw to breach Linux systems

1w ago · bleeping-computer

Hackers are actively exploiting a critical vulnerability in OpenSSH, dubbed 'SSHole' (CVE-2026-123158), to bypass authentication and gain unauthorized access to Linux systems, particularly those based on musl, such as Alpine Linux. The flaw allows attackers to log in to accounts with a password hash set to '*' by sending an oversized password, which triggers an error in musl's password hashing function that OpenSSH misinterprets as a valid match. Exploitation began shortly after public disclosure and proof-of-concept release, with confirmed attacks involving system reconnaissance, credential theft, persistence via SSH authorized_keys modification, and deployment of the TAC Agent Python-based implant.

3 IoCs 1 CVEs
Binance denies data breach after alleged ShinyHunters claims

1w ago · bleeping-computer

A threat actor using the 'ShinyHunters' alias claimed to have breached Binance and offered for sale data allegedly containing KYC documents, passwords, private keys, and transaction history of 335 million users. Binance denied the claims, stating there was no evidence of a breach and that the claims were likely a scam, possibly based on infostealer malware data from users' devices rather than a direct compromise. The actor demanded $900,000, but inconsistencies in the claim and skepticism from forum users cast doubt on its legitimacy. Separately, Binance recently disclosed a prior incident involving North Korean hackers from the Lazarus Group compromising an employee via a fake job interview, leading to a $6.34 billion cryptocurrency theft.

3 Actors
Russian cloud firm Beget rebuilds from scratch after data wiper attack

1w ago · bleeping-computer

Russian cloud services provider Beget suffered a destructive cyberattack on November 1, 2026, in which an attacker gained access to its management systems and deployed data-wiping malware across its infrastructure. The attack erased data from VPS, S3, Kubernetes, and CDN configurations, with only partial backups surviving. Beget is rebuilding its systems from scratch, treating all infrastructure as potentially compromised, and has not found evidence of data exfiltration—indicating the attack was purely destructive in nature.

Next →