Tag: cloud security
420 articles
AgentCore Harness Exposes Credential Risks
A default-enabled shell in AgentCore Harness can be exploited through prompt injection, allowing attackers to execute commands and extract plaintext credentials from the harness process memory, putting sensitive data at risk. This simple yet consequential chain highlights a critical vulnerability in credential security.
Gyazo Breach Exposes 23.6 Million User Records After Server Flaw Exploited
A massive data breach at Gyazo has exposed a staggering 23.6 million user records, thanks to a server vulnerability that was exploited by attackers on September 11, 2026. The incident has prompted the company to temporarily suspend its service for maintenance, and users are advised to wait patiently for its recovery.
Microsoft Fixes Critical Azure AI Foundry Flaw Enabling Privilege Escalation
Microsoft just patched a critical flaw in Azure AI Foundry that allowed unauthorized attackers to escalate privileges over a network - a vulnerability that earned a perfect 10.0 CVSS score for its severity. Thankfully, there's no evidence it's been exploited in the wild... yet.
Security Teams Struggle to Connect Dots in Cross-Environment Attacks
Attacks are getting more complex, with 43% of incidents spreading across four or more environments - and in some cases, as many as eight. Security teams struggle to piece together these cross-environment attacks, making it crucial to connect the dots between disparate signals.
Docker Flaw Lets Guest Code Read, Modify macOS Host Files
A newly discovered Docker flaw on macOS could allow malicious code in a virtual machine to break free from its sandbox and read or modify sensitive host files, potentially leading to code execution on the host. This vulnerability, tracked as CVE-2026-77179, leverages a weakness in the virtio-fs host server to gain unauthorized access.
AI Agents Expose New Attack Vectors
A shocking 230 out of 243 unauthenticated LocalAI instances were found to be vulnerable to exploitation, highlighting a disturbing trend: our increasing reliance on convenient tools is creating new avenues for old security mistakes. This alarming vulnerability was recently exploited in a large-scale campaign, compromising high-value targets, including infrastructure linked to the Thai military.
Brevo Cloudflare API Key Compromise Injects Malicious ClickFix Scripts
Attackers compromised a Cloudflare API key with full account permissions, allowing them to inject malicious scripts into website responses by creating a rogue Cloudflare Worker that cleverly evaded standard security checks. By modifying content at the CDN edge and removing key security headers, the hackers kept origin servers and files untouched, making the tampering difficult to detect.
City Relay Breach Exposes London Property Managers' Financial Data
London property managers City Relay have suffered a data breach, exposing sensitive financial information, after hackers exploited a vulnerability in a third-party cloud platform connected to their systems. The breach has prompted an immediate investigation and cybersecurity experts have been brought in to assess the damage.
Test Environment Exposes Live Customer Data Due to Oversight
A mid-size company's security audit uncovered a shocking oversight: a test environment was accidentally exposing live customer data to the internet, highlighting how easily security problems can arise from simple mistakes. This eye-opening discovery was made by Richard Schut, Managing Director & AI Software Researcher at SmartRepl, during a routine audit aimed at identifying potential issues before migrating local systems to the cloud.
CISA and NIST Issue Guidance to Safeguard Cloud Identity Tokens
As cyber threats escalate, protecting cloud identity tokens has become crucial - and CISA and NIST have just released guidance to help safeguard these vulnerable targets. The new interagency report provides critical advice for federal agencies, cloud service providers, and organizations that rely on them to stay secure.
Malware Exploits Chinese Casino Sites for Espionage
Uncover the dark side of Chinese-language casino sites: a vast network of seemingly harmless online casinos and adult sites is secretly harboring malware, facilitating espionage, and laundering money - with ties to major US cloud providers. These sites cleverly disguise themselves as legitimate entertainment hubs, using common templates and relying on the house edge to turn a profit.
Mass-Scanning Campaign Exploits Vite Flaw to Harvest Cloud Credentials
A high-severity flaw in Vite, tracked as CVE-2026-39364, has been exploited in a mass-scanning campaign, allowing attackers to bypass security restrictions and harvest cloud credentials. By manipulating query parameters, unauthenticated attackers can leak sensitive files that were meant to be blocked.
Cloud Identities Exposed Through Behavioral Clustering Analysis
Mapping the complex roles of human, machine, and autonomous agent identities in cloud environments is a daunting security challenge - but a practical, data-driven solution can turn audit logs into a revealing behavioral map. By analyzing API activity across 125 cloud environments, researchers have developed a clever method for grouping identities into functional roles, shining a light on what they actually do.
Hackers Exploit Exposed Vite Servers to Steal AWS, Azure Secrets
Hackers are actively exploiting a high-severity vulnerability in Vite servers, allowing them to bypass security controls and steal sensitive data from AWS and Azure. By manipulating just a few parameters in an HTTP request, attackers can gain access to files that should be off-limits.
Defense Cyber Spending Poised to Double by 2031
Get ready for a surge in defense cyber spending - it's expected to nearly double from $14.99 billion in 2026 to a whopping $28.75 billion by 2031. This dramatic growth is driven by increasing threats, a reliance on connected tech, and a new focus on cyber warfare capabilities.
Microsoft Cloud Accounts Targeted in Passkey Phishing Campaigns
Microsoft uncovered a massive phishing campaign that sent over a million scam emails in just three days, cleverly impersonating CEOs and other executives to trick companies into divulging sensitive info. The attackers then switched tactics, using passkey-themed social engineering to hijack cloud accounts and steal corporate files.
Anthropic Exposes AI Abuse by Hackers Linked to Russia, China
Anthropic uncovered a massive AI-powered operation by hackers linked to Russia and China, who used a sophisticated pipeline to download and scan 1.8 million Android apps for sensitive secrets. This alarming breach highlights the growing threat of AI abuse in cyber attacks.
Artifactory Flaws Exploited to Deploy Rust Backdoor Malware
Security researchers have uncovered a sneaky attack that uses two flaws in JFrog Artifactory to deploy a custom Rust backdoor malware, giving hackers full control of self-hosted servers. This stealthy exploit chain lets attackers escalate from low privileges to admin control, even when anonymous access is disabled.
AI Deployments Outpace Permissions Reviews, Leaving Organizations Exposed
Most organizations are launching AI tools without properly assessing the risks, leaving them exposed to potential security breaches. A recent survey found that while 76% of organizations have deployed AI tools, only 43% have completed a thorough review of permissions.
Attackers Exploit JFrog Flaws to Seize Admin Control
Cyber attackers have found a way to chain two JFrog Artifactory flaws, CVE-2026-42018 and CVE-2026-42016, to gain administrator control of self-hosted instances, putting your sensitive data at risk. This alarming exploit was recently reported by cloud security company Wiz.
IDScan Breach Exposes 153 Million Driver's Licenses
A massive data breach at IDScan has exposed a staggering 153 million driver's license scans, which were allegedly advertised for sale on a dark-web forum, compromising sensitive information on a huge scale. The identity verification firm has confirmed unauthorized access to its cloud platform and is working with specialists to investigate the incident.
LiteLLM Gateways Expose Cloud Credentials to Risk with Default Admin Key
Thousands of LiteLLM gateways are leaving cloud credentials vulnerable due to a shocking security flaw: nearly 300 online gateways accepted a default admin key, granting hackers unrestricted access to sensitive data and cloud accounts. This easily exploitable weakness puts countless organizations at risk of devastating breaches.
Slim Spider Targets Brazilian Finance with Cloud Credential Heists
Meet Slim Spider, a cunning cyber threat actor with a laser-like focus on Brazilian finance, expertly infiltrating cloud environments to steal sensitive credentials and target cryptocurrency assets and instant payment accounts. With a sophisticated grasp of Brazil's financial infrastructure, this adversary has been wreaking havoc on the country's financial sector since at least March 2026.
Threat Actors Exploit AI for Large-Scale Credential Harvesting
Threat actors are supercharging their attacks with AI, using it to scale up credential harvesting and launch more sophisticated campaigns. Google Threat Intelligence Group reports that financially motivated and espionage-focused attackers are increasingly turning to AI to target high-value sectors like healthcare, government, and media.