Lucene search
+L

7 matches found

RedhatCVE
RedhatCVE
added 2026/06/05 7:20 p.m.15 views

CVE-2026-41713

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns...

8.2CVSS5.4AI score0.00218EPSS
SaveExploits0References1
EUVD
EUVD
added 2026/05/12 12:32 p.m.14 views

EUVD-2026-29449

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns...

8.2CVSS5.7AI score0.00218EPSS
SaveExploits0References3
NVD
NVD
added 2026/05/12 11:16 a.m.17 views

CVE-2026-41713

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns...

8.2CVSS0.00218EPSS
SaveExploits0References2
ATTACKERKB
ATTACKERKB
added 2026/05/12 10:17 a.m.13 views

CVE-2026-41713

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns...

8.2CVSS5.7AI score0.00218EPSS
SaveExploits0References3Affected Software1
OSV
OSV
added 2026/05/12 10:17 a.m.7 views

CVE-2026-41713 Prompt Injection via Memory Poisoning in PromptChatMemoryAdvisor

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns...

8.2CVSS5.8AI score
SaveExploits0References4
Positive Technologies
Positive Technologies
added 2026/05/12 12:0 a.m.20 views

PT-2026-40007

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns...

8.2CVSS5.7AI score0.00218EPSS
SaveExploits0References3
Schneier on Security
Schneier on Security
added 2025/07/25 11:10 a.m.10 views

Subliminal Learning in AIs

Today's freaky LLM behavior: We study subliminal learning, a surprising phenomenon where language models learn traits from model-generated data that is semantically unrelated to those traits. For example, a "student" model learns to prefer owls when trained on sequences of numbers generated by a...

7.3AI score
SaveExploits0
Rows per page
Query Builder