Security news.
Today's cybersecurity landscape is dominated by active exploitation of critical vulnerabilities, with CISA confirming ransomware gangs are leveraging the "BlueHammer" flaw in Microsoft Defender. AI systems are also under significant attack, facing new research detailing methods to leak credentials and bypass safeguards, alongside widespread API key exposures in iOS apps.
CISA: Windows BlueHammer flaw now exploited by ransomware gangs
CISA has confirmed that ransomware groups are actively exploiting CVE-2026-33825, a Microsoft Defender privilege escalation vulnerability, which was previously abused as a zero-day.
Langflow RCE Exploited to Deploy Monero Miner on Exposed AI App Endpoints
Threat actors are actively exploiting CVE-2026-33017, an unauthenticated remote code execution (RCE) vulnerability in Langflow (CVSS 9.3), to deploy Monero cryptocurrency miners on exposed AI application endpoints.
Exploitation of Recent Oracle E-Business Suite Vulnerability Begins
A critical-severity vulnerability in Oracle E-Business Suite's Payments product is now being actively exploited, allowing unauthenticated attackers to take over the system.
Critical SimpleHelp Vulnerability Exploited for Malware Delivery
A critical authentication bypass vulnerability (CVE-2026-48558, CVSS 10.0) in SimpleHelp's OpenID Connect (OIDC) flow is being exploited to deliver new malware families, TaskWeaver and Djinn Stealer, targeting credentials.
282 iOS AI Apps Leak API Keys and Open AI Proxy Access in Network Traffic Study
A study of 444 iPhone AI chatbot apps found that 282 (nearly two-thirds) exposed paid AI access, often via plaintext API keys or reusable tokens in network traffic.
New BioShocking Attack Tricks AI Browsers Into Leaking User Credentials
The "BioShocking" technique can trick AI browsers and assistants, including ChatGPT Atlas and Perplexity's Comet, into copying and sending user credentials to an attacker by convincing them they are playing a game.
Aflac Japan Data Breach Impacts 4.38 Million
Aflac Japan disclosed a data breach affecting 4.38 million individuals after hackers accessed its policyholder portal multiple times between June 15 and June 25.
Microsoft Warns Poisoned MCP Tool Descriptions Can Make AI Agents Leak Data
Microsoft research demonstrates how attackers can hijack AI agents using poisoned tool descriptions to make them quietly exfiltrate company data, with the agent's actions appearing routine and avoiding alarms.