Bot Wars Evolved: Orchestrating Competing LLMs in a Counterstrike Against Phone Scams
Nardine Basta, Conor Atkins, Dali Kaafar · 2025
Read Bot Wars Evolved: Orchestrating Competing LLMs in a Counterstrike Against Phone Scams by Nardine Basta, Conor Atkins, Dali Kaafar (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-07036.txt.
Description
Nardine Basta, Conor Atkins, and Dali Kaafar turn large language models loose on phone scammers and let the models invent their own counters. Their Bot Wars framework fields scam-baiter bots that hold fake victims' conversations with fraudsters. A two-layer prompt design builds demographically believable personas while a chain-of-thought process grows strategy without explicit training for it. The test set holds 3,200 scam dialogues checked against 179 hours of human baiting.
Judged on cognitive, numeric, and content measures, GPT-4 sounded the most natural and true to persona. Deepseek kept conversations alive the longest. The setup formalizes how baiting tactics can emerge from reasoning alone.
Catalog record
| CREATOR | Nardine Basta, Conor Atkins, Dali Kaafar |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-07036