🔥 New Win‑DoS Flaws Turn DCs Into DDoS Botnets
SafeBreach Labs unveiled Win‑DoS, an RPC/LDAP-based assault method that abuses public Domain Controllers for massive distributed denial-of-service attacks. Four DoS flaws plus a novel zero-click DDoS chain elevate domain controllers to dangerous botnet assets.
Source: https://thehackernews.com/2025/08/new-win-ddos-flaws-let-attackers-turn.html Wikipedia+15The Hacker News+15Cyber Security News+15
🔐 Windows EPM Exploit Enables Privilege Escalation via RPC Poisoning
Researchers delivered a new exploit chain for the Windows RPC Endpoint Mapper (EPM). CVE‑2025‑49760 allows unprivileged users to impersonate core services, force authentication to malicious servers, and escalate privileges.
Source: https://thehackernews.com/2025/08/researchers-detail-windows-epm.html heise online+11The Hacker News+11X (formerly Twitter)+11
🤖 GPT‑5 Jailbreak Techniques Hit the Wild
New techniques demonstrate how GPT‑5 can be “jailbroken”—circumventing content restrictions for unintended behaviors. This signals increased urgency in securing AI models from prompt injection and context exploitation.
Source: https://thehackernews.com/2025/08/researchers-uncover-gpt-5-jailbreak-and.html
🎯 Revault Attacks Let Firmware Survive Reinstalls on Dell Laptops
The newly detailed ReVault attack targets ControlVault3 firmware on Dell laptops, offering persistence even after clean OS reinstalls, as well as physical bypass of Windows login.
Source: https://thehackernews.com/2025/08/researchers-reveal-revault-attack.html Onsite Computing, Inc.+11The Hacker News+11X (formerly Twitter)+11
📡 Scattered Spider Launches Telegram Breach Channel
The ransomware group has set up a new Telegram channel combining ShinyHunters and Lapsus$ branding—where they’re publicly posting attack claims and proof, all in real time.
Source: https://databreaches.net/2025/08/09/scattered-spider-has-a-new-telegram-channel-to-list-its-attacks/ DataBreaches.Net
Emerging Patterns to Watch
- Controller spaces and infrastructure layers are becoming covert botnets—DCs and RPC services included.
- Malicious logic is living deeper—firmware-level hacks like ReVault are difficult to detect or remediate.
- AI containment is slipping—with generative models losing their boundaries faster than defenses catch up.
- Threat actors are colocating with their victim communities—with public channels amplifying both reach and risk.