A New York Times analysis of seven million words spoken by President Donald Trump found that his public speeches have become more profane, more absolutist, longer-winded and less coherent over the past decade. The project combined artificial intelligence with traditional reporting and followed a similar study published in 2024.
Journalists Peter Baker and Dylan Freedman used AI to help identify examples of negative language, swearwords and rambling passages in Trump’s speeches and rallies. Freedman translated Baker’s questions into code and prompts designed to find relevant excerpts.
For the earlier project, Freedman compiled a 120-page document of excerpts from nine years of speeches using criteria they had developed. Both journalists reviewed the material, and selected examples were included in the story.
The reporters did not treat the AI’s results as verified findings. They checked each excerpt in context and also used basic statistical analysis of word frequencies to assess the range of Trump’s vocabulary.
Sources: The New York Times Politics.