This year's Pulitzer Prizes marked a record number of winners who disclosed the use of AI tools, with eight awardees revealing their reliance on artificial intelligence. The disclosures include five winners and three finalists, according to Nieman Lab. The requirement to disclose AI use has been in place since 2024, and this year saw a higher frequency of AI tool and large language model usage among entrants. The tools were mainly used to expedite the search through large document sets.
The Wall Street Journal used an internal LLM to summarize thousands of public documents related to Texas floods. The Minnesota Star Tribune used ChatGPT to translate a female shooter's diary written in faux Cyrillic, then had language experts verify the results. The Associated Press used an LLM to analyze tens of thousands of leaked documents on Chinese surveillance technology, while the New York Times used GPT-5 to verify its manual classification of SEC crypto cases. These examples highlight the varied applications of AI in journalistic processes.
Pulitzer administrator Marjorie Miller stated that the news industry now accepts 'AI is here to stay' and has a better understanding of appropriate use cases. She emphasized that the use of AI is acceptable in areas other than writing and editing stories that might be considered for a Pulitzer Prize. The disclosure rule will also extend to book entries starting next year.
Source: thedecoder
