Lucene search
K

6 matches found

Packet Storm News
Packet Storm News
added 2026/05/20 12:0 a.m.3 views

Backchaining Loss of Control Mitigations from Mission-Specific Benchmarks in National Security

Affordances and permissions are promising and timely safety levers for mitigating Loss of Control LoC threats in high-stakes deployment contexts, such as national security. Deployers in defense and intelligence could rely on several approaches to identify which affordances and permissions should ...

5.8AI score
Exploits0
Packet Storm News
Packet Storm News
added 2026/03/30 12:0 a.m.1 views

Why Aggregate Accuracy Is Inadequate for Evaluating Fairness in Law Enforcement Facial Recognition Systems

Facial recognition systems are increasingly deployed in law enforcement and security contexts, where algorithmic decisions can carry significant societal consequences. Despite high reported accuracy, growing evidence demonstrates that such systems often exhibit uneven performance across demograph...

5.9AI score
Exploits0
Packet Storm News
Packet Storm News
added 2026/03/16 12:0 a.m.1 views

Evasive Intelligence: Lessons from Malware Analysis for Evaluating AI Agents

Artificial intelligence AI systems are increasingly adopted as tool-using agents that can plan, observe their environment, and take actions over extended time periods. This evolution challenges current evaluation practices where the AI models are tested in restricted, fully observable settings. I...

5.8AI score
Exploits0
Packet Storm News
Packet Storm News
added 2026/02/23 12:0 a.m.4 views

LLM-Enabled Applications Require System-Level Threat Monitoring

LLM-enabled applications are rapidly reshaping the software ecosystem by using large language models as core reasoning components for complex task execution. This paradigm shift, however, introduces fundamentally new reliability challenges and significantly expands the security attack surface, du...

5.9AI score
Exploits0
Packet Storm News
Packet Storm News
added 2025/12/12 12:0 a.m.2 views

Persistent Backdoor Attacks under Continual Fine-Tuning of LLMs

Backdoor attacks embed malicious behaviors into Large Language Models LLMs, enabling adversaries to trigger harmful outputs or bypass safety controls. However, the persistence of the implanted backdoors under user-driven post-deployment continual fine-tuning has been rarely examined. Most prior...

7.2AI score
Exploits0
Snyk
Snyk
added 2025/08/28 1:33 p.m.1 views

Use of Default Credentials

Overview Affected versions of this package are vulnerable to Use of Default Credentials for the admin account. An attacker can gain full administrative access by using the default credentials if the password is not changed after deployment. Workaround This vulnerability can be mitigated by loggin...

9.8CVSS7.2AI score0.00098EPSS
Exploits0References2
Rows per page
Query Builder