orcarouter/OrcaSAQ-2-27B
Text Generation • 7B • Updated • 2.73k • 236
Ibrahim.H
bitsnaps
AI & ML interests
Text mining, sentiment analysis, text summarization
Recent Activity
liked a model about 8 hours ago
orcarouter/OrcaSAQ-2-27B liked a model about 16 hours ago
Edge0/Audio8-ASR-Infinite liked a model 2 days ago
MiniMaxAI/MiniMax-M3-MXFP8Organizations
liked a
model about 8 hours ago
liked a
model about 16 hours ago
liked a
model 2 days ago
reacted to mihailgribov's post with 👍 2 days ago
Post
1612
Will your AI agent tell you it was attacked?
We took the same agent from our earlier experiment and added one thing: a twentieth tool,
The system prompt said nothing about attacks or when to use it. We then ran the same 395 injected emails through nine agentic models.
Alarm rates ranged from 49% to zero.
The unexpected result came from the newest model in the test,
Astra did not follow a single injected payment instruction. But it did not report a single one either. On clean and injected emails alike, it simply read the email, logged the subject, and finished.
That is a useful distinction: resisting an attack and recognizing it as a security event are not the same capability.
A model can be perfectly resistant in this test and still leave you with no evidence that anyone attacked it.
Full experiment and results:
https://huggingface.co/blog/mihailgribov/will-the-agent-tell-you-it-was-attacked
Quadrat-IPI dataset:
mihailgribov/quadrat-ipi
Run your own model:
https://github.com/mihail-gribov/quadrat-ipi-model-eval
#prompt-injection #indirect-prompt-injection #agentic-ai #llm-security #ai-agents
We took the same agent from our earlier experiment and added one thing: a twentieth tool,
escalate_security_incident.The system prompt said nothing about attacks or when to use it. We then ran the same 395 injected emails through nine agentic models.
Alarm rates ranged from 49% to zero.
The unexpected result came from the newest model in the test,
gpt-6-astra.Astra did not follow a single injected payment instruction. But it did not report a single one either. On clean and injected emails alike, it simply read the email, logged the subject, and finished.
That is a useful distinction: resisting an attack and recognizing it as a security event are not the same capability.
A model can be perfectly resistant in this test and still leave you with no evidence that anyone attacked it.
Full experiment and results:
https://huggingface.co/blog/mihailgribov/will-the-agent-tell-you-it-was-attacked
Quadrat-IPI dataset:
mihailgribov/quadrat-ipi
Run your own model:
https://github.com/mihail-gribov/quadrat-ipi-model-eval
#prompt-injection #indirect-prompt-injection #agentic-ai #llm-security #ai-agents
- 7 replies
liked a
Space 3 days ago
liked a
dataset 3 days ago
liked a
model 3 days ago
liked a
Space 4 days ago
liked a
model 4 days ago
liked 3
models 5 days ago
liked 4
models 6 days ago
updated a
Space 6 days ago
published a
Space 6 days ago
updated a
dataset 6 days ago
published a
dataset 6 days ago