Researchers at the Bioinformatics Open Source Conference (BOSC) for 2026 developed an AI agent called bosc-pre-review to assist human reviewers in evaluating open-source software submissions. This agent assessed abstracts based on criteria such as openness, licensing, and ease of execution, with a tool named Runabilly testing project buildability and safety in Docker containers. While the AI provided evidence for reviewers, human decision-making remained paramount. Post-review surveys indicated that most reviewers found the AI's assistance useful, though they preferred to independently verify the AI's conclusions rather than blindly accept them. AI
IMPACT This research demonstrates a practical application of AI in streamlining the peer-review process for open-source software, potentially improving efficiency for conference organizers and reviewers.
RANK_REASON The item describes an experience report from a conference detailing the use of an AI agent for pre-reviewing open-source software submissions, which falls under research into AI applications. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
