How publisher Wiley is leaning into AI

When we think of AI in healthcare, our minds tend to go to tech companies like Abridge or Nvidia. But what about the research that teaches large language models (LLMs) about medicine?

In early March, “ChatGPT for doctors” OpenEvidence announced it had added academic journal and research publisher Wiley’s content to its platform, which provides real-time answers based on peer-reviewed research to queries from medical professionals. OpenEvidence also partnered with journals JAMA Network and the New England Journal of Medicine in June 2025 and February 2026, respectively, with this deal marking a continued shift that may prompt medical researchers to update their strategies in the face of AI.

With a 219-year history, Wiley publishes 1,500+ peer-reviewed journals and 1,500+ books each year, according to its website. In 2026, that work will also mean incorporating new technologies, like AI.

“We wanted to ultimately be able to shape how AI influenced our ecosystem, rather than be shaped by it,” Josh Jarrett, Wiley’s SVP and general manager of AI growth, told us.

Leaning in. When ChatGPT came on the scene about three years ago, Jarrett said he realized AI would “have a big impact on publishing and our work, whether we like it or not.”

“It probably is going to create risks, too, and we’re going to be best served by leaning in and trying to experiment, explore, build relationships, learn as much as we can, and then by participating in that ecosystem,” he added.

This approach has influenced a few big decisions. Wiley started by making its archival content available to train models—like OpenEvidence’s LLM—including published articles, textbooks, and other course materials, as well as for real-time searches.

“It turns out things that are good for humans to learn are also good for machines,” Jarrett said.

In a 2025 study of 2,430 researchers, Wiley found that 84% reported they are using AI for their work, up from 57% the year prior. The researchers do this through subscription knowledge feeds that allow them to access, say, oncology research. Now, information that has historically been available solely to industry specialists can now be accessed by everyday users in a more digestible manner, Jarrett said.

“We have this great opportunity with AI to exponentially expand the number of people who could benefit from the insights and the evidence and the powerful content, and find patterns in there, find unknown insights,” Jarrett said.

Wiley is also working to build more partnerships like the one it has with OpenEvidence. On March 5, the publisher linked up with Amazon Web Services and Microsoft’s Dragon Copilot to help speed up literature searches.

These partnerships and additional users also impact Wiley’s customer base and business strategy.

“It enables us to reach huge new audiences,” Jarrett said.

Ultimately, he said the company is committed to providing quality information to people seeking medical information. When they use AI, they should know whether information is coming from a Reddit post or a trusted medical journal.

“Enabling that ecosystem of trust, accuracy, and attribution—we take [that] very seriously,” Jarrett said.