The ability of Artificial Intelligence to read and summarize text is already making it a useful tool for scholarship. How will it change the stories we tell about the past? Bill Wasik, New York Times Magazine’s editorial director has some thoughts ('A.I. is Poised to Rewrite History. Literally', NYTMadgazine June 16, 2025).
He points out that in the digital age, historians face an embarrassment of riches: more digitized sources than they could read in a lifetime. From JSTOR to Google Books, the sheer volume of searchable archives has changed the way we approach the past. Now, artificial intelligence promises to take that transformation even further, offering tools that can scan, summarize, and synthesize enormous amounts of text in seconds. But as with every technological leap in the organization of knowledge, from Callimachus’s “Pinakes” to the Dewey Decimal System, efficiency comes with trade-offs.
On one hand, the rise of large language models (LLMs) presents tantalizing opportunities. Why spend months sifting through sources when A.I. can do it in minutes? For nonfiction writers and historians overwhelmed by the deluge of information, there’s undeniable appeal in outsourcing the burden of reading. Scholars are already experimenting with these tools, and the idea of a future where a chatbot assists in synthesizing material across languages, continents, and centuries is no longer far-fetched.
But lurking behind the promise is a familiar danger: the flattening of nuance. Summarization, one of A.I.’s superpowers, risks turning textured, contradictory historical accounts into palatable, seemingly coherent narratives. Historian Lara Putnam has warned that digital search already encourages “finding without knowing where to look,” untethering researchers from the place-specific context that traditionally grounded their work. Artificial Intelligence could amplify that disconnect, replacing deep archival immersion with plausible-sounding syntheses that obscure as much as they reveal.
There’s also the risk that digitized and AI-parsed history reflects a narrow, biased archive — skewed toward English, elite institutions, and official records. If we increasingly see the past through what is “digitizable”, Wasik argues that we may lose sight of those who’ve always lived at history’s margins.
Still, innovation marches on. Some developers envision e-books enhanced by embedded A.I., allowing readers to ask follow-up questions, explore original sources, or view timelines generated on demand. In this vision, the historian’s work isn’t a static text but a living, interactive experience; a book with a built-in research assistant. It’s a staggering possibility. Whether it heralds a utopia of democratized knowledge or a dystopia of oversimplified truths remains uncertain.
In the end, the key question may not be whether A.I. helps or hurts the writing of history, but what kind of history we want to write, and who we trust to write it.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment