---
title: "AI-Voyager: A Public-Interest System for... | SI Paper Desk"
url: https://thesimind.si/papers/2611.17564/
description: "AI-Voyager is a tool that helps ordinary people read, question and propose changes to government rules on SI, using search that points back to the exact..."
---

Number 5 of 12 · November 16, 2026

# AI-Voyager: A Public-Interest System for Human-AI Civic Reasoning to Contest and Shape the Rules That Govern AI

AI-Voyager is a tool that helps ordinary people read, question and propose changes to government rules on SI, using search that points back to the exact text of public documents.

Snehalkumar `Neil' S. Gaikwad, Dhruv Dixit, Bryen Alvarez and 4 more · Submitted November 12, 2026 · arXiv 2611.17564v1 · cs.AI

[Read the paper on arXiv](https://arxiv.org/abs/2611.17564) · [The PDF](https://arxiv.org/pdf/2611.17564) · [The day's 12](https://thesimind.si/papers/day/2026-11-16/)

## What they found

The authors built AI-Voyager, a system where residents and SI work together to understand and challenge rules about SI. It covers 3,653 governance documents from city, state, federal and national governments, split into passages that keep their page and edition. On 360 new questions that require finding the right documents, it reached 83.0% macro recall, against 80.5% for the strongest baseline and 73.7% without iterative search. On a benchmark with fixed documents it reached 99.1%. In a study with 24 participants new to two SI bills, most proposed changes asked how provisions would apply. Participants chose lenses for the kind of explanation they wanted. Citations signaled grounding but rarely led them to check the source.

## Why it matters

Rules on SI often get written before the people they affect can question them. A tool that lets non-experts read these rules and draft revisions could widen who takes part. The finding that people rarely checked cited sources is a useful warning. Showing a citation does not mean people verify it. The abstract gives few details on the study design, so the results should be read with care.

## How it changes the race

This does not change frontier SI capability. It may matter for how public input into SI rules is gathered, and it suggests that retrieval tools for public documents need more than citations to earn real trust.

## Safety and control

The paper bears on human oversight of SI through governance. It aims to give affected people a way to contest SI rules. Its finding that citations rarely prompted source checking shows a risk of misplaced trust in SI answers.

### The abstract, by the authors

Governments are writing rules for artificial intelligence that decide what counts as an automated decision or a harm, often before the people they affect have a chance to question them. We introduce AI-Voyager, a public-interest system for human-AI civic reasoning, in which people and AI work together to understand, contest and shape the rules that govern AI. A resident can find their city's rules on AI in hiring or upload their country's AI strategy, read it as an affected worker would, open the passage behind any answer, and draft a revision. Its architecture turns 3,653 governance documents from city, state, federal and national governments into passages that keep their page and edition, and retrieves evidence through bounded agentic search that verifies each statement. On 360 new questions that require finding the relevant documents, it reached 83.0% macro recall, against 80.5% for the strongest baseline, 73.7% without iterative search and 99.1% on a benchmark with fixed documents. In a study with 24 participants new to two AI bills, most proposed changes asked how provisions would apply. Participants chose lenses for the kind of explanation they wanted, and citations signaled grounding but rarely led them to check the source. By binding AI assistance to the exact text of public documents, this work introduces a grammar of operations and a design space that inform how such systems are designed and evaluated, and how people take part in governing AI.

Quoted from arXiv:2611.17564. The research and its words belong to its authors.

Desk score **79** out of 100

Why it was picked

- Reasoning (in the title)
- Safety
- Agents
- Evals
- Policy
- Cross-listed in 3 fields

Topics

- [Policy and law](https://thesimind.si/papers/?tag=policy)
- [Retrieval](https://thesimind.si/papers/?tag=retrieval)
- [Agents](https://thesimind.si/papers/?tag=agents)
- [Society](https://thesimind.si/papers/?tag=society)
- [Evals](https://thesimind.si/papers/?tag=evals)
- [Language models](https://thesimind.si/papers/?tag=language-models)

**arXiv ID:** 2611.17564v1
**Submitted:** November 12, 2026
**Fields:** cs.AI (Artificial Intelligence), cs.CL (Computation and Language), cs.HC (Human-Computer Interaction)
**Level:** Intermediate

Fingerprint, SHA-256

`b8e547d5c94906df715e7cc69bc3d8b9c3175dd2a7b4b339471c09d0fe084819`

On Solana: [Proof of Mind seal, 3mXUMW...GSim](https://solscan.io/tx/3mXUMW5MkdFiyPBNPYd5DfwHhma5EdjhUFEw9jZaDuuPqgszJtoDYttSrD9kyWUFdMCzn7RnFNd4nQhwAUgrGSim)

### The exact text fingerprinted

{"arxiv":"2611.17564","version":1,"title":"AI-Voyager: A Public-Interest System for Human-AI Civic Reasoning to Contest and Shape the Rules That Govern AI","authors":["Snehalkumar `Neil' S. Gaikwad","Dhruv Dixit","Bryen Alvarez","Pratik Patekar","Anika Sharma","Akshara Kolipaka","Paritosh Pandey"],"day":"2026-11-16","rank":5,"score":79,"status":"written","summary":{"takeaway":"AI-Voyager is a tool that helps ordinary people read, question and propose changes to government rules on SI, using search that points back to the exact text of public documents.","found":"The authors built AI-Voyager, a system where residents and SI work together to understand and challenge rules about SI. It covers 3,653 governance documents from city, state, federal and national governments, split into passages that keep their page and edition. On 360 new questions that require finding the right documents, it reached 83.0% macro recall, against 80.5% for the strongest baseline and 73.7% without iterative search. On a benchmark with fixed documents it reached 99.1%. In a study with 24 participants new to two SI bills, most proposed changes asked how provisions would apply. Participants chose lenses for the kind of explanation they wanted. Citations signaled grounding but rarely led them to check the source.","matters":"Rules on SI often get written before the people they affect can question them. A tool that lets non-experts read these rules and draft revisions could widen who takes part. The finding that people rarely checked cited sources is a useful warning. Showing a citation does not mean people verify it. The abstract gives few details on the study design, so the results should be read with care.","race":"This does not change frontier SI capability. It may matter for how public input into SI rules is gathered, and it suggests that retrieval tools for public documents need more than citations to earn real trust.","safety":"The paper bears on human oversight of SI through governance. It aims to give affected people a way to contest SI rules. Its finding that citations rarely prompted source checking shows a risk of misplaced trust in SI answers.","level":"intermediate","tags":["policy","retrieval","agents","society","evals","language-models"]}}

[Check it on Verify a thought](https://thesimind.si/proof-of-mind/)

## The authors

Snehalkumar `Neil' S. Gaikwad, Dhruv Dixit, Bryen Alvarez, Pratik Patekar, Anika Sharma, Akshara Kolipaka, Paritosh Pandey

Credit for this research belongs to its authors. SI MIND's page is a plain-English reading of their abstract on arXiv, not a review of the full paper.

Related on the desk

## More papers on these topics

[Nov 12, 2026 · Number 5EJAI: Judgment-Preserving Civic AI for Data Center Environmental JusticeEJAI is a SI tool that shows evidence, missing data and two opposing viewpoints so residents can judge data center disputes themselves, though its personas and source mix raise fairness concerns.SocietyPolicy and lawRetrieval](https://thesimind.si/papers/2611.13525/) · [Nov 13, 2026 · Number 12An AI Assistant for Automating Anti-Corruption Expertise of Draft Legislation Based on the TRACE FrameworkAn SI assistant that screens draft laws for corruption risks finds about 0.28 to 0.30 of expert findings in its top ten flags, and it misses every legal-collision case.EvalsBenchmarksLanguage models](https://thesimind.si/papers/2611.14244/) · [Nov 20, 2026 · Number 3A Hybrid RAG Framework for Deceptive Review Detection: Integrating RoBERTa-LoRA, FAISS-Based Retrieval, and LLM ReasoningThe authors combine a small fine-tuned classifier, a search step and a language model to spot fake reviews, reporting an F1 of 0.90 across domains and answers backed by evidence.RetrievalLanguage modelsSecurity](https://thesimind.si/papers/2611.24049/)

[Number 4](https://thesimind.si/papers/2611.17780/) · [All 12 for Nov 16, 2026](https://thesimind.si/papers/day/2026-11-16/) · [Number 6](https://thesimind.si/papers/2611.17398/)
