10. Data Poisoning / Model Drift via Auto-Learning
10. Data Poisoning / Model Drift via Auto-Learning
If MCP agents fine-tune on user input without filtering, malicious prompt sequences can shift behavior over time. Impact: Gradual normalization of insecure responses or biased outputs. Mitigation:- No auto-training on live, unsanitized data
- Add poisoned input detection + human-in-the-loop review
Uncategorized @us-en
Oct 16, 2026
CVE-2026-102764: BeyondTrust Zero-Day Exploited
BeyondTrust has disclosed CVE-2026-102764, a critical Remote Support vulnerability exploited as a zero-day weeks before customers received a patch. The attack began…
Cyber News
Oct 15, 2026
Cisco Secure Email & PAN-OS Zero-Days Exploited
Cisco Secure Email & PAN-OS Zero-Days Exploited Cisco and Palo Alto Networks have patched two actively exploited zero-day vulnerabilities that can lead…
CTI Fundamentals
Oct 15, 2026
SOCRadar Splunk Integration: New MSSP and Vulnerability Apps
SOCRadar Splunk Integration: New MSSP and Vulnerability Apps Your analysts live in Splunk. Your external risk does not: SOCRadar alarms for dozens…
Intelligence in your inbox every Thursday.
Join 30,000+ security professionals reading SOCRadar research.
Subscribe