Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Publishers Data topic
No spam. Unsubscribe anytime.
Witnesses urge Congress to examine publishers' data products and AI risks in scholarly publishing
Summary
Academic witnesses warned that major publishers now collect and monetize sensitive research and personnel data and that AI tools can be used to fabricate plausible but false papers; they recommended treating research data as a public asset and building accountable infrastructure.
Dr. Jason Owen Smith told the subcommittee that large publishers increasingly behave like data analytics firms, collecting detailed information about how articles are read, reviewed and used and sometimes acquiring tools used in faculty hiring and evaluation. "I can't tell you what Elsevier knows about me and my work," he said, describing concerns about data access, privacy and foreign data sharing.
Members pressed witnesses on national‑security and compliance implications. Dr. Owen Smith pointed to public privacy‑policy language from vendor sites indicating possible data‑sharing partners abroad and argued Congress should consider a regulatory framework that treats data about federally funded research as a strategic asset while investing in publicly accountable infrastructure to manage it.
Witnesses also discussed the growing role of artificial intelligence in creating fraudulent or low‑quality manuscripts. They said AI lowers the technical cost of producing plausible‑looking papers, and that detecting such content requires combining textual analysis with citation and usage metadata. The panel urged more funding for both investigatory offices and technical detection tools, and asked publishers to provide documentation about their data engines and governance for the record.
