eLife Pathways is eLife's open-publishing-infrastructure initiative (backed by a £2.4m Wellcome investment), building on its publish-review-curate model toward modular publishing — separating a paper into parts so both humans and machines can link between them, with different composed views for different audiences. As part of it, eLife has worked with a group using LLMs to extract claims, evidence and hypotheses from papers in structured form, and is prototyping an "AI-scientist sandbox" that reasons over all extracted claims at once to surface new hypotheses and adversarial AI peer review.
Benefits & outcomes
In a 10-paper pilot, structured claim extraction surfaced two labs working on the same neuroscience target without citing or knowing about each other.
Modular, machine-linkable outputs enable large-scale pattern-finding across methods and reproducibility.
Attempts to recreate the computational side of papers, exposing gaps where code or data is missing.
Notes & quotes
Described by Paul Shannon (Head of Technology & Innovation, eLife) at PurePub.ai's "Journal 2030" panel (21 May 2026). Related tooling discussed: Stencila and the nascent OXA standard for interoperability between Stencila, Myst and Markdown notebooks. Wellcome funding: £2.4m for eLife Pathways.
Publisher / company: eLife
Date of mention: 21/05/2026
Source: https://elifesciences.org/