Anthropic Warns Infostealer Malware Hijacking Claude Sessions
September 2, 2026
Anthropic warns Vidar, Lumma, StealC, RedLine, and AMOS malware are stealing Claude session tokens, enabling attackers to drain user credits
Endpoint Security
Anthropic Warns Infostealer Malware Hijacking Claude Sessions
Anthropic warns Vidar, Lumma, StealC, RedLine, and AMOS malware are stealing Claude session tokens, enabling attackers to drain user credits fraudulently.
AI Research Org METR Loses $600K in Credits to Dual Breach Attacks
METR disclosed two incidents where attackers stole API keys and consumed $600,000 in AI credits through fail-open authentication and targeted probing.
Russia-Aligned UAC-0099 Embeds Nuclear Weapon Text to Evade AI Tools
Russian threat actor UAC-0099 deployed GuardBreaker technique, inserting safety-sensitive phrases into malware to trip AI security analysis mechanisms.
Five Venezuelan Nationals Plead Guilty to Kansas ATM Jackpotting
Five Venezuelan nationals pleaded guilty to ATM jackpotting conspiracy following December 2025 arrests for Tren de Aragua malware operations in Kansas.
JFrog Artifactory CVE-2026-82329 Exploited Days After Disclosure
WatchTowr researchers observed attackers exploiting CVE-2026-82329 to mint admin tokens on JFrog Artifactory instances days after August 28 disclosure.
TerminalFix Campaign Uses Reverse Tunnels in ClickFix-Style Attacks
TerminalFix deploys multistage PowerShell attacks incorporating reverse tunnels into victim networks, using ClickFix social engineering to trick users.
Sandworm Fake Job Interview Campaign Targets Ukrainian IT Workers
CERT-UA attributes a Sandworm-linked UAC-0145 social engineering campaign using fake job interviews and trojanized WireGuard VPN clients against Ukraine.
Kimwolf v7 Android Botnet Evades DDoS Mitigation Using HTTP/2 C2
Palo Alto Unit 42 documents Kimwolf v7 using HTTP/2 C2 to mimic legitimate browsing, evade DDoS detection, and expand across Android and IoT devices worldwide.
Zapscape KVM Flaw Lets Privileged L1 Guest Escape to Host
Researcher Hyunwoo Kim documented Zapscape, CVE-2026-64561, a KVM/x86 shadow memory flaw allowing privileged L1 guest code to escape to the Linux host.
TONTOU Interrupt Injection Bypasses Spectre v2 Fixes on AMD Zen 2
MIT CSAIL's interrupt injection attack named TONTOU bypasses retpoline and Safe-RET defenses on AMD Zen 2 to leak Linux password hashes from userspace.
AI Research Org METR Loses $600K in Credits to Dual Breach Attacks
September 1, 2026
METR disclosed two incidents where attackers stole API keys and consumed $600,000 in AI credits through fail-open authentication and targeted
Russia-Aligned UAC-0099 Embeds Nuclear Weapon Text to Evade AI Tools
September 1, 2026
Russian threat actor UAC-0099 deployed GuardBreaker technique, inserting safety-sensitive phrases into malware to trip AI security analysis mechanisms.
Five Venezuelan Nationals Plead Guilty to Kansas ATM Jackpotting
September 1, 2026
Five Venezuelan nationals pleaded guilty to ATM jackpotting conspiracy following December 2025 arrests for Tren de Aragua malware operations in
JFrog Artifactory CVE-2026-82329 Exploited Days After Disclosure
September 1, 2026
WatchTowr researchers observed attackers exploiting CVE-2026-82329 to mint admin tokens on JFrog Artifactory instances days after August 28 disclosure.
TerminalFix Campaign Uses Reverse Tunnels in ClickFix-Style Attacks
September 1, 2026
TerminalFix deploys multistage PowerShell attacks incorporating reverse tunnels into victim networks, using ClickFix social engineering to trick users.
Sandworm Fake Job Interview Campaign Targets Ukrainian IT Workers
August 12, 2026
CERT-UA attributes a Sandworm-linked UAC-0145 social engineering campaign using fake job interviews and trojanized WireGuard VPN clients against Ukraine.
Kimwolf v7 Android Botnet Evades DDoS Mitigation Using HTTP/2 C2
August 12, 2026
Palo Alto Unit 42 documents Kimwolf v7 using HTTP/2 C2 to mimic legitimate browsing, evade DDoS detection, and expand across
Zapscape KVM Flaw Lets Privileged L1 Guest Escape to Host
August 7, 2026
Researcher Hyunwoo Kim documented Zapscape, CVE-2026-64561, a KVM/x86 shadow memory flaw allowing privileged L1 guest code to escape to the
TONTOU Interrupt Injection Bypasses Spectre v2 Fixes on AMD Zen 2
August 7, 2026
MIT CSAIL’s interrupt injection attack named TONTOU bypasses retpoline and Safe-RET defenses on AMD Zen 2 to leak Linux password
Trending
Daily Briefing Newsletter
Subscribe to the Daily Security Review Email Briefing to stay informed on the latest threats, trends, and technology, along with insightful columns from industry experts.