Meta disclosed that more than 20,000 Instagram users had their accounts hijacked after attackers abused its AI-powered High Touch Support (HTS) recovery system to obtain password reset links. The company has disabled the flawed tool, reset affected accounts, and said it will fix the email verification process before relaunching the service. #Meta #Instagram #HighTouchSupport #HTS
Category: Cyber Security News
Microsoft has released an open-source fork of Windows Terminal called Intelligent Terminal, which brings AI assistance directly into the terminal without interrupting the regular shell session. It can explain errors, draft commands, remember past agent sessions, and resume previous work through integrations such as GitHub Copilot, Claude, Codex, and Gemini. #WindowsTerminal #IntelligentTerminal #GitHubCopilot #Claude #Codex #Gemini
The Silent Ransom Group is targeting U.S. law firms and professional services organizations with invoice-themed phishing, impersonation of IT staff, and remote support abuse to steal sensitive data within hours. Mandiant and the FBI say the gang, also tracked as UNC3753, Luna Moth, and Chatty Spider, uses aggressive extortion tactics and fast-moving infrastructure to pressure victims into paying. #SilentRansomGroup #UNC3753 #LunaMoth #ChattySpider #Mandiant #FBI
Fortinet researchers uncovered C0XMO, a new and more modular Gafgyt botnet variant that targets DD-WRT routers and can spread across many CPU architectures and device types. It exploits CVE-2021-27137, brute-forces weak SSH/Telnet credentials, and supports a wide range of DDoS capabilities while also hunting down competing tools and botnet clients on infected hosts. #C0XMO #Gafgyt #DDWRT #CVE202127137
Daily Recap, Actively exploited flaws were updated across major products, with CISA adding the SolarWinds Serv-U DoS issue to KEV despite no patch for CVE-2026-20245 and also flagging issues in Cisco Catalyst SD-WAN Manager. Malware and ecosystem threats continued alongside browser- and web-based activity, including Chinese APT persistence tooling, Android spyware Asin targeting Arabic users, and supply-chain abuse via IronWorm and Miasma on npm. #SolarWinds #Serv-U #CVE-2026-20245 #Cisco #CatalystSD-WAN #KEV #FFmpeg #Asin #IronWorm #Miasma #npm #OP-512 #MicrosoftIIS #Toshiba #Muji #Max
Emphere, a Seattle-based startup, raised $2.1 million in pre-seed funding to build an AI-driven vulnerability remediation platform that analyzes software dependency graphs and automatically applies safe fixes. The company aims to help software teams remediate vulnerabilities faster across modern code stacks without breaking downstream systems. #Emphere #AI2Incubator #OutsidersFund…
Hackers are actively exploiting CVE-2026-3300 in the Everest Forms Pro plugin to gain unauthenticated code execution and create rogue administrator accounts on WordPress sites. Wordfence says attacks began on April 13, with more than 29,300 blocked attempts and indicators tied to IPs 202.56.2[.]126 and 209.146.60.26. #CVE-2026-3300 #EverestFormsPro #Wordfence #diksimarina
OpenAI is rolling out Lockdown Mode for eligible ChatGPT personal and Business accounts to reduce data exfiltration risks from prompt injection attacks. The feature limits web and external-service access by disabling tools like live browsing, image retrieval, deep research, Agent mode, Canvas networking, and file downloads. #OpenAI #ChatGPT #LockdownMode #PromptInjection…
Daily Recap, Agentic AI and enterprise identity security are in focus as experts warn that autonomous systems can amplify insider risk, while developer tooling flaws and browser updates add new supply-chain and patching pressure. Multiple high-impact breaches and theft activity were reported across organizations including DentaQuest, RCI, and the UN World Food Programme, alongside new incidents like the IronWorm npm supply-chain attack and the Hola Browser cryptominer delivery. #AgenticAI #EnterpriseIdentity #ClaudeCode #GitHubActions #DentaQuest #RCI #UNWorldFoodProgramme #WFP #Stripe #IronWorm #npm #HolaBrowser #Chrome149 #Cisco #SDWAN #UnifiedCM #CVE202620230 #Comodo #FiveEyes #ChineseSpies #Kremlin #FTC #X #CISA #Palantir #Anthropic #BraveOrigin
An autonomous AI agent found 21 previously unknown vulnerabilities in FFmpeg, including long-dormant issues, while Google released Chrome 149 with patches for a record 429 security bugs. The surge in AI-assisted findings is increasing the pace and volume of vulnerability disclosure across projects like FFmpeg and Chrome, forcing faster triage and…
CISA has added CVE-2026-28318, a high-severity denial-of-service flaw in SolarWinds Serv-U, to its Known Exploited Vulnerabilities catalog after evidence of active exploitation. SolarWinds has released a fix in version 15.5.4 HF1 and recommends limiting access and blocking requests containing “content-encoding” while FCEB agencies must remediate by June 19, 2026. #SolarWinds #Serv-U…
A researcher reverse-engineered Bright Dataβs iOS SDK and found it can turn consumer devices, including always-on smart TVs, into exit nodes for web-scraping traffic. The findings highlight a consent gap in Bright Dataβs proxy model, where home IPs and bandwidth are used for AI-driven scraping through devices that may bypass VPNs…
Opal Security has raised $23 million in new funding, bringing its total investment to $59 million, to expand its AI-native identity governance platform. The companyβs platform helps organizations control access for employees, service accounts, and AI agents across cloud, SaaS, and on-premises environments. #OpalSecurity #Greylock #BatteryVentures #CambiumCapital…
Cyberattackers are targeting Internet-exposed automatic tank gauge (ATG) systems in the United States, prompting a joint warning from multiple federal agencies. Attackers could alter tank readings or disable alerts, so organizations are urged to remove ATGs from the open Web, patch systems, and strengthen access controls. #CISA #FBI #NSA #ATG
Cisco has warned that CVE-2026-20245, a high-severity flaw in Cisco Catalyst SD-WAN Manager, is being actively exploited and can let an authenticated local attacker execute arbitrary commands as root by uploading a crafted file. The issue affects multiple SD-WAN deployment types and has already been observed causing configuration changes on edge…