conv.

All stories
AIQuiet 10d · day 16

Nature paper unveils Paper2Agent, turning research papers into AI agents

A new framework converts papers into MCP-based AI agents that reproduce results and answer new queries, sparking praise and skepticism online.

What to know

  • Paper2Agent automatically builds an MCP-based AI agent from a paper's text, code and data, letting users query it in natural language via tools like Claude Code.
  • In a case study, the AlphaGenome paper agent was built in ~45 minutes for $14 and outscored other biomedical AI agents, including Biomni, on genetics questions.
  • The agent surfaced a different candidate causal gene for a cholesterol-linked variant than the original paper, which the authors say shows the tool can help re-evaluate published conclusions.
  • Online reaction splits between researchers calling the paper 'incredible' and skeptics questioning whether papers should require an AI intermediary to be understood.

The dispute Whether converting papers into AI-queryable agents solves a real accessibility problem or undermines the value of readers actually understanding research themselves. · positions read across 3 posts and comments

some voices

The framework is an exciting advance that will change how scientific knowledge is disseminated and reused.

  • “Nature has just published a pretty incredible paper. The authors introduce a framework that turns research papers into interactive AI agents.”

    Valerio Capraro · LinkedIn (via Mastodon repost) ↗
some voices

The premise that papers create barriers requiring 'understanding' misses that comprehension is the whole point of reading research.

  • “I didn't get past that sentence. READERS TO UNDERSTAND. Isn't that the point?”

    KarenKeiller@cosocial.ca · Mastodon ↗

“Conventional research papers require readers to understand and adapt the paper's code, data and methods to their work, creating barriers to dissemination and reuse.”

Miao, Davis, Zhang, Pritchard & Zou, Paper authors · Nature ↗ · Sep 15

James Zou Co-author; computer scientist, Stanford UniversityJiacheng Miao Lead co-author of the Paper2Agent paperValerio Capraro LinkedIn commenterBiomni Rival academic biomedical AI agent tool

Nature paper unveils Paper2Agent, turning research papers into AI agents
nature.com

How it unfolded 4 developments, newest first · click a bar or a number to jump articlesposts

Peak 4 pieces in 4h at Sep 15, 7 PM; 20 pieces over 16 days (12 articles · 8 posts) Sep 12, 7 AM — 1 piece · 1 post — Hacker News 1Sep 12, 11 AM — quietSep 12, 3 PM — quietSep 12, 7 PM — quietSep 12, 11 PM — quietSep 13, 3 AM — quietSep 13, 7 AM — quietSep 13, 11 AM — quietSep 13, 3 PM — quietSep 13, 7 PM — quietSep 13, 11 PM — quietSep 14, 3 AM — quietSep 14, 7 AM — quietSep 14, 11 AM — quietSep 14, 3 PM — quietSep 14, 7 PM — quietSep 14, 11 PM — quietSep 15, 3 AM — quietSep 15, 7 AM — quietSep 15, 11 AM — quietSep 15, 3 PM — quietSep 15, 7 PM — 4 pieces · 3 articles · 1 post — Newswires 3, Mastodon 1Sep 15, 11 PM — quietSep 16, 3 AM — quietSep 16, 7 AM — quietSep 16, 11 AM — 3 pieces · 3 articles — Google News 2, Newswires 1Sep 16, 3 PM — 3 pieces · 3 articles — Google News 3Sep 16, 7 PM — 4 pieces · 4 posts — Mastodon 3, Reddit 1Sep 16, 11 PM — 1 piece · 1 article — Google News 1Sep 17, 3 AM — quietSep 17, 7 AM — 2 pieces · 1 article · 1 post — Google News 1, Hacker News 1Sep 17, 11 AM — quietSep 17, 3 PM — 1 piece · 1 post — Hacker News 1Sep 17, 7 PM — quietSep 17, 11 PM — 1 piece · 1 article — Newswires 1Sep 18, 3 AM — quietSep 18, 7 AM — quietSep 18, 11 AM — quietSep 18, 3 PM — quietSep 18, 7 PM — quietSep 18, 11 PM — quietSep 19, 3 AM — quietSep 19, 7 AM — quietSep 19, 11 AM — quietSep 19, 3 PM — quietSep 19, 7 PM — quietSep 19, 11 PM — quietSep 20, 3 AM — quietSep 20, 7 AM — quietSep 20, 11 AM — quietSep 20, 3 PM — quietSep 20, 7 PM — quietSep 20, 11 PM — quietSep 21, 3 AM — quietSep 21, 7 AM — quietSep 21, 11 AM — quietSep 21, 3 PM — quietSep 21, 7 PM — quietSep 21, 11 PM — quietSep 22, 3 AM — quietSep 22, 7 AM — quietSep 22, 11 AM — quietSep 22, 3 PM — quietSep 22, 7 PM — quietSep 22, 11 PM — quietSep 23, 3 AM — quietSep 23, 7 AM — quietSep 23, 11 AM — quietSep 23, 3 PM — quietSep 23, 7 PM — quietSep 23, 11 PM — quietSep 24, 3 AM — quietSep 24, 7 AM — quietSep 24, 11 AM — quietSep 24, 3 PM — quietSep 24, 7 PM — quietSep 24, 11 PM — quietSep 25, 3 AM — quietSep 25, 7 AM — quietSep 25, 11 AM — quietSep 25, 3 PM — quietSep 25, 7 PM — quietSep 25, 11 PM — quietYesterday, 3 AM — quietYesterday, 7 AM — quietYesterday, 11 AM — quietYesterday, 3 PM — quietYesterday, 7 PM — quietYesterday, 11 PM — quietToday, 3 AM — quietToday, 7 AM — quietToday, 11 AM — quietToday, 3 PM — quietToday, 7 PM — quiet 12–4
Sep 14Sep 16Sep 18Sep 20Sep 22Sep 24now · 9:47 PM ET
  1. 4

    Academic commenter questions the tool's premise

    A Mastodon user objected to the paper's framing that papers create barriers requiring readers to 'understand' them, arguing comprehension is the point of reading a paper.

    “READERS TO UNDERSTAND. Isn't that the point?”
    — KarenKeiller@cosocial.ca
    • KarenKeiller@cosocial.ca

      “Conventional research papers require readers to understand and adapt the paper’s code, data and methods to their work, creating barriers to dissemination and reuse.” I didn't get past that sentence. READERS TO UNDERSTAND. Isn't that the point? # academicchatter # scholcomm https://www. nature.com/articles/s41586-026 -11044-y

      KarenKeiller@cosocial.caMastodon10d ago1▲view on Mastodon ↗
  2. 3

    Valerio Capraro calls the paper 'pretty incredible' on LinkedIn

    A LinkedIn post praising the framework as turning papers into interactive AI agents was cross-posted to Mastodon and drew 38 comments.

    “Nature has just published a pretty incredible paper. The authors introduce a framework that turns research papers into interactive AI agents.”
    — Valerio Capraro
    1. first by Nature News, 12d ago

    • nic221@techhub.social

      Nature has just published a pretty incredible paper. The authors introduce a framework that turns research papers into interactive AI agents. Rather than engaging with a paper only as static text… | Valerio Capraro | 38 comments https://www. linkedin.com/posts/valerio-cap…

      nic221@techhub.socialMastodon11d agoview on Mastodon ↗
    2 posts in this stretch →
  3. 2

    Story reaches Reddit

    A Reddit user reposted the Nature News article on Paper2Agent, extending the story's reach beyond academic and science-media channels.

  4. background

    Story spreads via cross-posts on Mastodon and Flipboard — Automated and bridged accounts (Flipboard's NatureNewsteam, bsky.brid.gy bridges) reposted the Nature News piece across Mastodon instances shortly after publication.

  5. background

    Companion Nature pieces summarize the tool for broader audiences — Nature published a short explainer and a features piece describing Paper2Agent as a 'virtual corresponding author' that makes papers easier to reproduce, reuse and extend.

  6. background

    Nature News details the AlphaGenome case study — Nature's coverage explains that Paper2Agent built an agent for the AlphaGenome paper in about 45 minutes for $14, that it outscored other biomedical AI agents including Biomni, and that it flagged a different causal gene than the original paper for a cholesterol-linked variant.

  7. 1

    Nature publishes the Paper2Agent framework

    Miao, Davis, Zhang, Pritchard and Zou describe Paper2Agent, which converts manuscripts, code and data into MCP-based tool-invoking AI agents that reproduce original results and answer new queries.

    “Paper2Agent introduces a paradigm for knowledge dissemination and a collaborative ecosystem of AI co-scientists.”
    — Miao, Davis, Zhang, Pritchard & Zou
    1. first by Nature, 11d ago · also Yahoo Tech, Inshorts, Nature News

      2 more headlines

Also covered reported alongside — the timeline has no entry for these yet

  1. first by AI Weekly, 10d ago · also Marginal Revolution, Nature

    1 more headline
  2. first by Forkast, 11d ago · also MarkTechPost

    1 more headline

What people are saying 0 voices from 0 sites · best of 3 · verbatim