Xtasis Launches Tool to Verify Research Citations

CHICAGO, Sept. 22, 2026 /CourierPR/ -- New capability from Xtasis Inc., unveiled at AWS Imagine, is designed to help researchers, reviewers, publishers, and research teams validate cited references at scale. The company, which is behind CiteGeist, an AI-powered reference manager and research platform, has announced the launch of CiteGeist Reference Verification, a new feature designed to help researchers, reviewers, publishers, and research teams identify fabricated, mismatched, incomplete, or otherwise problematic references before publication. The release of CiteGeist Reference Verification is part of Xtasis' broader work on AI systems that emphasize transparency and traceability.

The growing use of generative AI in research and publishing has introduced new questions about how cited references are generated and validated. While AI can support literature review and manuscript preparation, generated citations may contain inaccurate bibliographic information, misattributed sources, invalid links, or even fabricated references. Evidence suggests that fabricated references are becoming more common. A 2026 audit published in The Lancet, analyzing 2.5 million biomedical papers over a three-year period, found that the rate of fabricated references remained relatively stable at approximately 4 per 10,000 papers throughout 2023 before beginning to rise sharply in mid-2024. The rate then increased quarter over quarter, reaching approximately 57 per 10,000 papers by early 2026. The authors also noted that reference verification is not routinely incorporated into the review process.

Xtasis Inc. has addressed this issue with CiteGeist Reference Verification, a tool that runs on Amazon Web Services (AWS) and incorporates NVIDIA Nemotron 3 into its AI technology stack. Reference Verification extends CiteGeist's existing Reference Management capabilities across literature discovery, organization, reading, annotation, synthesis, and citation generation and formatting. Ahead of the public launch, the tool was developed and refined with input from reviewers, publishers, and research teams, including users working with large reference sets. Their feedback helped shape the verification workflow around how references are reviewed in practice. The tool is designed to move reference validation earlier in the research process, allowing users to review references individually or in bulk, flag those that require closer examination, and generate a shareable verification report for coauthors, reviewers, publishers, and editorial teams.

Every reference matters, because even a single inaccurate or fabricated citation can undermine the integrity of the work. But reference integrity also becomes a workflow challenge at scale. A reviewer may need to assess hundreds of references, a publisher may be evaluating hundreds of manuscripts, and a research team may be synthesizing an entire body of literature. Xtasis Inc. wanted to make rigorous reference verification practical and manageable in each of those settings.

The launch of CiteGeist Reference Verification reflects a broader shift in how AI-enabled research tools may need to evolve: not only helping researchers work faster, but also making it easier to verify the information those workflows produce and rely on. CiteGeist Reference Verification is now available through CiteGeist.io.

Xtasis Inc. develops AI technologies for healthcare, research, and other data-intensive environments. Its products include CiteGeist™, a research and reference management platform; PixelGuard™, de-identification software for DICOM and other medical visual information; and NoteGuard™, de-identification software for clinical text and other unstructured data.