Search papers, labs, and topics across Lattice.
Affiliation:
2
0
5
LLM agents in decentralized marketplaces are demonstrably exploitable, with adversarial prompting enabling over 50% fraudulent transaction completion and doubling agent profits.
System prompts in commercial AI products are often a mixed bag, with 40% harboring instructions that can undermine user interests, revealing a critical gap in accountability.
