Timezone: »

Exploiting Programmatic Behavior of LLMs: Dual-Use Through Standard Security Attacks
Daniel Kang · Xuechen Li · Ion Stoica · Carlos Guestrin · Matei Zaharia · Tatsunori Hashimoto
Event URL: https://openreview.net/forum?id=eXwzgiXYM8 »

Recent advances in instruction-following large language models (LLMs) have led to dramatic improvements in a range of NLP tasks. Unfortunately, we find that the same improved capabilities amplify the dual-use risks for malicious purposes of these models. Dual-use is difficult to prevent as instruction-following capabilities now enable standard attacks from computer security. The capabilities of these instruction-following LLMs provide strong economic incentives for dual-use by malicious actors. In particular, weshow that instruction-following LLMs can produce targeted malicious content, including hate speech and scams, bypassing in-the-wild defenses implemented by LLM API vendors. Our analysis shows that this content can be generated economically and at cost likely lower than with human effort alone. Together, our findings suggest that LLMs will increasingly attract more sophisticated adversaries and attacks, and addressing these attacks may require new approaches to mitigations.

Author Information

Daniel Kang (Stanford University)
Xuechen Li (Stanford University)
Ion Stoica (University of California, Berkeley)
Carlos Guestrin (Stanford University & Apple)
Matei Zaharia (Stanford University)
Tatsunori Hashimoto (Stanford)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors