TY  - UNPD
A1  - Bauer, Kevin
A1  - Liebich, Lena
A1  - Hinz, Oliver
A1  - Kosfeld, Michael
T1  - Decoding GPT’s hidden 'rationality' of cooperation
N2  - In current discussions on large language models (LLMs) such as GPT, understanding their ability to emulate facets of human intelligence stands central. Using behavioral economic paradigms and structural models, we investigate GPT’s cooperativeness in human interactions and assess its rational goal-oriented behavior. We discover that GPT cooperates more than humans and has overly optimistic expectations about human cooperation. Intriguingly, additional analyses reveal that GPT’s behavior isn’t random; it displays a level of goal-oriented rationality surpassing human counterparts. Our findings suggest that GPT hyper-rationally aims to maximize social welfare, coupled with a strive of self-preservation. Methodologically, our esearch highlights how structural models, typically employed to decipher human behavior, can illuminate the rationality and goal-orientation of LLMs. This opens a compelling path for future research into the intricate rationality of sophisticated, yet enigmatic artificial agents.
T3  - SAFE working paper - 401 
KW  - large language models
KW  - cooperation
KW  - goal orientation
KW  - economic rationality
Y1  - 2023
UR  - http://publikationen.ub.uni-frankfurt.de/frontdoor/index/index/docId/71519
UR  - https://nbn-resolving.org/urn:nbn:de:hebis:30:3-715196
UR  - https://ssrn.com/abstract=4576036
N1  - We gratefully acknowledge research support from the University of Mannheim, the Leibniz Institute for Financial Research SAFE, and the Goethe University Frankfurt.
PB  - SAFE
CY  - Frankfurt am Main
ER  -