How good is GPT at writing political speeches for the White House?

TL;DR

This study evaluates GPT's ability to generate political speeches by comparing its outputs to real US presidential addresses, revealing stylistic differences and limitations in mimicking authentic presidential rhetoric.

Contribution

It provides a comparative analysis of GPT-generated speeches versus actual presidential addresses, highlighting stylistic and tonal differences and assessing GPT's capacity to emulate political speech.

Findings

GPT overuses 'we' and produces shorter, longer sentences.

GPT tends to adopt an optimistic tone with more political and abstract terms.

Both GPT versions differ significantly from real presidential speeches.

Abstract

Using large language models (LLMs), computers are able to generate a written text in response to a us er request. As this pervasive technology can be applied in numerous contexts, this study analyses the written style of one LLM called GPT by comparing its generated speeches with those of the recent US presidents. To achieve this objective, the State of the Union (SOTU) addresses written by Reagan to Biden are contrasted to those produced by both GPT-3.5 and GPT-4.o versions. Compared to US presidents, GPT tends to overuse the lemma "we" and produce shorter messages with, on average, longer sentences. Moreover, GPT opts for an optimistic tone, opting more often for political (e.g., president, Congress), symbolic (e.g., freedom), and abstract terms (e.g., freedom). Even when imposing an author's style to GPT, the resulting speech remains distinct from addresses written by the target author. Finally, the two GPT versions present distinct characteristics, but both appear overall dissimilar to true presidential messages.