Two New York Times journalists analyzed roughly seven million words spoken by Donald Trump in public speeches. They combined artificial intelligence tools with traditional reporting methods to identify recurring patterns. The project examined a large body of remarks delivered over multiple years.
The analysis relied on A.I. to process the enormous volume of speech transcripts efficiently. Reporters then reviewed the findings using established journalistic practices. This hybrid approach allowed for both broad pattern detection and careful contextual verification.
Automated tools helped surface trends that would be difficult to catch through manual review alone. Human journalists checked the flagged passages for accuracy and meaning. The combination aimed to balance speed with editorial judgment.
The review focused on public speeches rather than private conversations or written statements. Researchers looked for shifts in language, repeated themes, and changes over time. The scope covered millions of words delivered across numerous events.
A.I. analysis can scan text for word frequency, phrasing patterns, and topic emphasis at scale. Traditional reporting adds interpretation and filters out misleading signals. The team used both to reduce the risk of drawing false conclusions.
The project reflects a growing trend of newsrooms adopting computational methods for large-scale text analysis. Such approaches let reporters examine material that would be impractical to review line by line. The work still depends on human oversight to ensure accuracy.
Findings from the analysis were used to support reporting on speech trends rather than to generate automated stories. The journalists treated A.I. output as a starting point for further inquiry. Final judgments remained with the reporters.
