arxiv: 2605.01525 · v1 · submitted 2026-05-02 · 💻 cs.CY

Recognition: unknown

Computational Challenges in Scaling Democratic Deliberation

Davide Grossi

Authors on Pith no claims yet

Pith reviewed 2026-05-09 17:46 UTC · model grok-4.3

classification 💻 cs.CY

keywords digital democracydeliberationcomputational challengesalgorithmic problemspreference aggregationconsensus mechanismsartificial intelligenceonline platforms

0 comments

The pith

Digital democracy software requires new algorithms to manage large-scale deliberation on proposals, preferences, and consensus.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper outlines the core functionalities needed for digital systems to support democratic deliberation at scale, such as handling proposals, aggregating opinions, and facilitating agreement among large groups. It claims these requirements generate novel computational challenges that demand algorithmic solutions from computer science and artificial intelligence. A sympathetic reader would care because effective tools could allow broader participation in democratic processes without losing coherence or fairness. The work begins constructing a structured inventory of these problems and situates them relative to existing research.

Core claim

Developing functionalities for digital democracy software to support deliberative processes at scale poses novel computational challenges that require algorithmic solutions to interesting mathematical problems, and the paper takes initial steps toward a structured inventory of such problems while positioning possible approaches within current computer science and artificial intelligence research.

What carries the argument

A structured inventory of computational challenges tied to the core functionalities of scaling deliberative processes, including proposal management, preference aggregation, and consensus mechanisms.

If this is right

Targeted research in AI and algorithms can address specific open problems in preference aggregation and consensus for large groups.
Solutions would enable practical digital platforms capable of handling deliberation beyond small-group limits.
The inventory provides a map that links these problems to established areas like computational social choice and automated negotiation.
Future software designs can incorporate the identified mathematical structures to improve scalability.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Solving these challenges could allow public consultations and citizen assemblies to operate at national or international scales with maintained quality.
The problems may intersect with existing work on multi-agent systems, suggesting hybrid approaches that combine deliberation tools with existing AI planners.
Empirical tests could compare deliberation outcomes in small versus large groups using prototype systems built around the inventory.

Load-bearing premise

The listed core functionalities form the right minimal set for scaling deliberation and the associated problems are sufficiently novel and unsolved in existing computer science and AI literature to justify a new inventory.

What would settle it

Showing that every listed functionality can already be implemented using standard, off-the-shelf algorithms from current literature without requiring new research would disprove the need for a fresh inventory of challenges.

Figures

Figures reproduced from arXiv: 2605.01525 by Davide Grossi.

**Figure 1.** Figure 1: The deliberation-support loop in online deliberation platforms.   A11 A12 ... A1m A21 A12 ... A2m . . . . . . . . . . . . An1 An2 ... Anm     ⋆ 1 ... ⋆ 0 0 ... ⋆ . . . . . . . . . . . . 1 ⋆ ... 0   view at source ↗

**Figure 2.** Figure 2: Attitudes matrix (left) for m ideas and n participants. In the simple approval/disapproval case (right), each entry Aip (with 1 ≤ i ≤ n and 1 ≤ p ≤ m) can get value 1 (approval), 0 (disapproval), or ⋆ (undefined). Independent inputs. All ideas are independent pieces of texts entered in the system. This is the case, for instance, in Polis. Interdependent inputs. Ideas may exhibit explicit interdependencies … view at source ↗

**Figure 3.** Figure 3: Plot of the scores that the functions of equations (2) (blue) and (3) (green) would count for a single participant when a given number ℓ (here in the range 0 to 5) of ideas they support is selected as representative. The two papers I mentioned earlier by Halpern et al. (2023) and Lindeboom et al. (2025) design and study algorithms that can efficiently approximate solutions to the computational problem I de… view at source ↗

read the original abstract

The paper provides an overview of core functionalities that digital democracy software needs to provide in order to support democratic deliberative processes at scale. Developing these functionalities poses novel computational challenges and requires algorithmic solutions to interesting mathematical problems. The aim of the paper is to break the first ground towards a structured inventory of such problems, and to position possible approaches to them within current academic research in computer science and artificial intelligence.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This is a clean overview paper that inventories open computational problems in scaling deliberation without new theorems or data.

read the letter

The main thing to know is that Grossi has written a positioning piece that lists core functionalities digital democracy tools would need at scale and flags the algorithmic gaps around them. It does not claim to close any of those gaps or run new experiments. What it does well is give a compact map of the terrain: it breaks deliberation support into concrete tasks like argument aggregation, preference elicitation under uncertainty, and moderation at volume, then points to relevant corners of computational social choice and AI where partial techniques already exist. That framing is useful for someone entering the area or looking for thesis topics. The soft spots are modest but real. The paper asserts that these challenges are novel enough to warrant a fresh inventory, yet it does not include a systematic comparison against the last decade of work on online deliberation platforms or multi-agent systems; a reader has to take the author's word that the listed problems sit outside current CS coverage. Because the piece is purely descriptive, there are no derivations or falsifiable claims to check, which keeps the circularity burden at zero but also limits how much weight the inventory can carry on its own. The assumptions about which functionalities are minimal and necessary are stated plainly rather than defended with evidence from deployed systems. This paper is for researchers already working in computational social choice or civic technology who want a quick orientation to open questions. A reader who needs concrete algorithms or empirical results will find little here. It is coherent on its own terms and shows clear thinking about the literature, so it deserves a serious referee rather than a desk reject. I would send it out for review with the expectation that referees will ask for tighter citations and perhaps a short section on why certain problems have not been solved by adjacent fields yet.

Referee Report

1 major / 2 minor

Summary. The manuscript provides an overview of core functionalities required for digital democracy software to support democratic deliberation at scale. It identifies associated computational challenges, argues that these require algorithmic solutions to interesting mathematical problems, and aims to break initial ground toward a structured inventory of such problems while positioning possible approaches within current computer science and artificial intelligence research.

Significance. If the inventory accurately captures key challenges and their connections to CS/AI, the paper could help direct research attention toward under-explored intersections between computational methods and democratic processes, such as scalable argument aggregation or secure large-scale participation mechanisms. As a purely descriptive positioning exercise without new derivations, data, or proofs, its value lies in synthesis and agenda-setting rather than advancing a specific technical result.

major comments (1)

The central claim that the listed functionalities pose 'novel' computational challenges (abstract and introduction) rests on domain knowledge rather than explicit contrasts with existing literature in computational social choice, argument mining, or multi-agent systems; without such contrasts the inventory's novelty and the need for a new structured list remain difficult to assess.

minor comments (2)

The paper would benefit from a dedicated related-work section or table that maps each identified functionality to specific prior papers or systems, even if only to note gaps.
Some functionality descriptions could be clarified with brief concrete examples drawn from existing deliberation platforms to illustrate the scale at which the computational issues arise.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the major comment below and will revise the manuscript accordingly.

read point-by-point responses

Referee: The central claim that the listed functionalities pose 'novel' computational challenges (abstract and introduction) rests on domain knowledge rather than explicit contrasts with existing literature in computational social choice, argument mining, or multi-agent systems; without such contrasts the inventory's novelty and the need for a new structured list remain difficult to assess.

Authors: We agree that the manuscript would be strengthened by more explicit contrasts with the literature to substantiate the novelty of the structured inventory. While the paper already positions the challenges within CS and AI research, it does not include detailed side-by-side comparisons. In the revised version, we will add a dedicated subsection (in the introduction or as part of Section 2) that directly contrasts our inventory with related work. For computational social choice, we will note that existing results on voting and preference aggregation typically address static or one-shot decisions, whereas scaling deliberation requires handling iterative, multi-turn argument structures with consensus and inclusivity constraints. In argument mining, we will reference NLP techniques for extraction and summarization but explain that they lack integration with fairness, security, and large-scale human participation mechanisms central to democratic processes. For multi-agent systems, we will highlight differences in modeling human agents with privacy and accessibility requirements versus purely artificial agents. These additions will clarify the unique combinations of challenges that motivate our inventory without claiming the underlying techniques are entirely new. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper is an overview and positioning exercise that inventories core functionalities and computational challenges for scaling democratic deliberation. It presents no derivations, equations, predictions, fitted parameters, or mathematical results that could reduce to inputs by construction. The central aim is explicitly to 'break the first ground towards a structured inventory' and position approaches within existing CS/AI research, with no self-referential loops, self-citation load-bearing claims, or ansatzes smuggled in. The argument is self-contained as a survey without any reduction of outputs to the paper's own definitions or prior self-citations.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper contains no mathematical derivations, fitted parameters, or new theoretical constructs; it draws on standard background in computer science and social choice without introducing free parameters, axioms beyond domain norms, or invented entities.

pith-pipeline@v0.9.0 · 5338 in / 966 out tokens · 35746 ms · 2026-05-09T17:46:11.567433+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

29 extracted references · 3 canonical work pages · 1 internal anchor

[1]

Bachmann, F., Sarasua, C., and Bernstein, A. (2024). Fast and adaptive questionnaires for voting advice applications. In Joint European Conference on machine learning and knowledge discovery in databases , pages 365--380. Springer

2024
[2]

Behrens, J., Kistner, A., Nitsche, A., and Swierczek, B. (2014). Principles of Liquid Feedback . Interaktieve Demokratie

2014
[3]

Bera, S., Chakrabarty, D., Flores, N., and Negahbani, M. (2019). Fair algorithms for clustering. Advances in Neural Information Processing Systems , 32

2019
[4]

Brandt, F., Conitzer, V., and Endriss, U. (2012). Computational social choice. Multiagent systems , 2:213--284

2012
[5]

W., Lopez-Lopez, E., Hechtlinger, S., Rahwan, Z., Aeschbach, S., Bakker, M

Burton, J. W., Lopez-Lopez, E., Hechtlinger, S., Rahwan, Z., Aeschbach, S., Bakker, M. A., Becker, J. A., Berditchevskaia, A., Berger, J., Brinkmann, L., et al. (2024). How large language models can reshape collective intelligence. Nature human behaviour , 8(9):1643--1655

2024
[6]

Chhabra, A., Masalkovait \.e , K., and Mohapatra, P. (2021). An overview of fairness in clustering. IEEE Access , 9:130698--130720

2021
[7]

Dahl, R. A. (2008). Polyarchy: Participation and opposition . Yale university press

2008
[8]

J., Comer, D

Denning, P. J., Comer, D. E., Gries, D., Mulder, M. C., Tucker, A., Turner, A. J., and Young, P. R. (1989). Computing as a discipline. Computer , 22(2):63--70

1989
[9]

o lz, P., Parkes, D. C., Procaccia, A. D., Rusak, G., Shapira, I., and W \

Fish, S., G \"o lz, P., Parkes, D. C., Procaccia, A. D., Rusak, G., Shapira, I., and W \"u thrich, M. (2024). Generative social choice. In Proceedings of the 25th ACM Conference on Economics and Computation , pages 985--985

2024
[10]

Grossi, D., Hahn, U., M \"a s, M., Nitsche, A., Behrens, J., Boehmer, N., Brill, M., Endriss, U., Grandi, U., Haret, A., et al. (2024). Enabling the digital democratic revival: A research program for digital democracy. arXiv preprint arXiv:2401.16863

work page arXiv 2024
[11]

D., Tucker-Foltz, J., and W \"u thrich, M

Halpern, D., Kehne, G., Procaccia, A. D., Tucker-Foltz, J., and W \"u thrich, M. (2023). Representation with incomplete votes. In Proceedings of the AAAI Conference on Artificial Intelligence , volume 37, pages 5657--5664

2023
[12]

and Peters, J

Kellerhals, L. and Peters, J. (2024). Proportional fairness in clustering: A social choice perspective. Advances in Neural Information Processing Systems , 37:111299--111317

2024
[13]

and Skowron, P

Lackner, M. and Skowron, P. (2022). Approval-based committee voting. In Multi-Winner Voting with Approval Preferences , pages 1--7. Springer

2022
[14]

and Reed, C

Lawrence, J. and Reed, C. (2019). Argument mining: A survey. Computational linguistics , 45(4):765--818

2019
[15]

Lederer, P. (2025). Proportional representation in rank aggregation. arXiv preprint arXiv:2508.16177

work page arXiv 2025
[16]

Lindeboom, F., Brehm, M., Grossi, D., and Murukannaiah, P. (2025). A voice for minorities: diversity in approval-based committee elections under incomplete or inaccurate information. arXiv preprint arXiv:2506.10843

work page internal anchor Pith review Pith/arXiv arXiv 2025
[17]

and Chwalisz, C

McKinney, S. and Chwalisz, C. (2025). Five dimensions of scaling democratic deliberation: With and beyond ai. Brussels: DemocracyNext. Retrieved April , 19:2025

2025
[18]

Mikhaylovskaya, A. (2024). Enhancing deliberation with digital democratic innovations. Philosophy & Technology , 37(1):3

2024
[19]

Narayanan, A. (2023). Understanding social media recommendation algorithms

2023
[20]

Rawls, J. (2017). A theory of justice. In Applied ethics , pages 21--29. Routledge

2017
[21]

Revel, M., Milli, S., Lu, T., Watson-Daniels, J., and Nickel, M. (2025). Representative ranking for deliberation in the public sphere. In Forty-second International Conference on Machine Learning

2025
[22]

M., Honemann, D

Robert III, H. M., Honemann, D. H., Balch, T. J., Seabold, D. E., and Gerber, S. (2020). Robert's rules of order newly revised . PublicAffairs

2020
[23]

A., et al

Rosa, P., Guimares, P. A., et al. (2025). Harnessing innovation on online deliberation. European Commission, Joint Research Center

2025
[24]

Salganik, M. J. and Levy, K. E. (2015). Wiki surveys: Open and quantifiable social data collection. PloS one , 10(5):e0123483

2015
[25]

Settles, B. (2009). Active learning literature survey

2009
[26]

Skowron, P., Lackner, M., Brill, M., Peters, D., and Elkind, E. (2017). Proportional rankings. In Proceedings of the 26th International Joint Conference on Artificial Intelligence , pages 409--415

2017
[27]

Small, C., Bjorkegren, M., Erkkil \"a , T., Shaw, L., and Megill, C. (2021). Polis: Scaling deliberation by mapping high dimensional opinion spaces. Recerca: revista de pensament i an \`a lisi , 26(2)

2021
[28]

S., Barto, A

Sutton, R. S., Barto, A. G., et al. (1998). Reinforcement learning: An introduction , volume 1. MIT press Cambridge

1998
[29]

H., Bakker, M

Tessler, M. H., Bakker, M. A., Jarrett, D., Sheahan, H., Chadwick, M. J., Koster, R., Evans, G., Campbell-Gillingham, L., Collins, T., Parkes, D. C., et al. (2024). Ai can help humans find common ground in democratic deliberation. Science , 386(6719):eadq2852

2024