Csoonline Single Prompt Disables Safety in 15 Language Models
Article Content
Browse articles 
Microsoft researchers revealed that a benign prompt can disable safety features in 15 major language models. The technique, known as GRP-Obliteration, exploits a training method called Group Relative Policy Optimization, which is typically used to enhance model safety. This discovery raises significant implications for AI alignment in enterprise applications.
Ask AI about this cluster
Answers cite the sources they use
Updated 241d ago How this analysis works
Timeline
2026-02-09
Microsoft published research detailing the prompt's effects
2026-02-10
CSO Online article published on the findings
More articles in this cluster (2)
Continue Reading
Volt Typhoon's Silent Train Network Targets US Critical Infrastructure The Chinese state-sponsored hacking group Volt Typhoon has been linked to a covert network called Silent Train, which has exploited at least four zero-day vulnerabilities over three years. This network utilizes rapidly rotating chains of compromised SOHO routers, IoT devices, and enterprise infrastructure to access… 
GitLab Issues Emergency Patch for Critical RCE Vulnerability GitLab has released emergency patches for a critical remote code execution vulnerability, CVE-2026-97229, affecting its Community and Enterprise Editions. This flaw, stemming from an integer overflow in the CI/CD regular expression compiler, allows unauthenticated attackers to execute arbitrary code via specially…