Cybersecurity News | Daily Recap [11 Sep 2026]

AI Threats

• Anthropic said Russian hackers used Claude to automate malware evasion, while its research also warned that AI can let smaller actors run state-level campaigns and a researcher resigned over AI safety concerns – Claude Abuse, AI Campaigns, AI Warning
• AI agents exploited PaperCut flaws to compromise 395 organizations, highlighting automated attack chains that combine vulnerability exploitation with scaling – PaperCut Attack, AI Exploit
• OpenAI faced Senate scrutiny over a Hugging Face breach probe as Cyber Command tapped an intelligence veteran for a top AI role – OpenAI Probe, AI Hire

Vulnerability Exploits

• GitLab urged users to patch a max-severity path traversal flaw, while Check Point disclosed two 9.8-rated VPN certificate bugs enabling unauthenticated RCE – GitLab Patch, VPN Flaws
• Cisco FMC flaws were reportedly exploited by a ransomware gang and state-sponsored hackers, and a new BlueMoon kit was found abusing Windows and Chrome zero-days – Cisco Exploits, BlueMoon Kit
• Microsoft shipped fixes for Teams and Outlook launch failures on ARM Windows PCs, while Windows Server September updates broke Remote Desktop Services and an Excel patch disrupted copy-paste for some users – Teams/Outlook Fix, RDS Breakage, Excel Bug

Malware & Mobile

• A new Android malware strain was seen encrypting files, stealing data, and harassing victims, while another roundup warned of 200 Android flaws, browser-built phishing, and 119K scam shops – Android Malware, ThreatsDay
• Google Play Early Access was abused to push thousands of deceptive Android apps, and passkeys can now move between password managers on Android – Play Abuse, Passkey Transfer
• Trezor said 347,000 users were targeted in phishing after the Brevo breach, showing how third-party incidents fuel downstream credential attacks – Trezor Phish

Data Breaches & Fraud

• IDScan confirmed a breach tied to 153 million stolen driver’s licenses offered on the dark web, underscoring the scale of identity-data exposure – IDScan Leak, IDScan Notice, IDScan Confirmed
• The Treasury urged banks to report cyber scams, citing nearly $13 billion in losses since 2023, and a report noted cheap SIM-swapping style attacks can still kick a stranger’s phone off the network – Treasury Scam, SIM Costs

Ransomware & Criminal Justice

• A Conti ransomware gang member was sentenced to 4 years in prison, while guidance on building a ransomware decision tree and reporting scam losses aimed to improve preparedness – Conti Sentence, Conti Conviction, Ransomware Prep

VPNs & Service Incidents

• Surfshark said hackers breached its internal testing and proxy servers, adding to vendor-security concerns around private network infrastructure – Surfshark Hit, Surfshark Breach

Industry Moves

• Kiteworks acquired Bonfy.AI to expand runtime data governance and close the AI governance gap, while Kevin Mandia joined the Amazon board and the month saw 33 cybersecurity M&A deals – Kiteworks Deal, Kiteworks Expand, Mandia Joins, M&A Roundup

Research & Risk

• Security research said companies may be measuring phishing resilience the wrong way, while another report highlighted the top threats found by investigating every alert for a quarter – Phishing Metrics, Threat Review
• ThreatsDay also surfaced broader ecosystem risks including Android flaws, browser phishing, and large-scale scam operations – ThreatsDay

Cybersecurity News | Daily Recap – hendryadrian.com