Wisdom of the Silicon Crowd: LLM Ensemble Prediction Capabilities Rival Human Crowd Accuracy


Paper by Philipp Schoenegger, Indre Tuminauskaite, Peter S. Park, and Philip E. Tetlock: “Human forecasting accuracy in practice relies on the ‘wisdom of the crowd’ effect, in which predictions about future events are significantly improved by aggregating across a crowd of individual forecasters. Past work on the forecasting ability of large language models (LLMs) suggests that frontier LLMs, as individual forecasters, underperform compared to the gold standard of a human crowd forecasting tournament aggregate. In Study 1, we expand this research by using an LLM ensemble approach consisting of a crowd of twelve LLMs. We compare the aggregated LLM predictions on 31 binary questions to that of a crowd of 925 human forecasters from a three-month forecasting tournament. Our preregistered main analysis shows that the LLM crowd outperforms a simple no-information benchmark and is not statistically different from the human crowd. In exploratory analyses, we find that these two approaches are equivalent with respect to medium-effect-size equivalence bounds. We also observe an acquiescence effect, with mean model predictions being significantly above 50%, despite an almost even split of positive and negative resolutions. Moreover, in Study 2, we test whether LLM predictions (of GPT-4 and Claude 2) can be improved by drawing on human cognitive output. We find that both models’ forecasting accuracy benefits from exposure to the median human prediction as information, improving accuracy by between 17% and 28%: though this leads to less accurate predictions than simply averaging human and machine forecasts. Our results suggest that LLMs can achieve forecasting accuracy rivaling that of human crowd forecasting tournaments: via the simple, practically applicable method of forecast aggregation. This replicates the ‘wisdom of the crowd’ effect for LLMs, and opens up their use for a variety of applications throughout society…(More)”.

Citizen Engagement in Evidence-informed Policy-making: A Guide to Mini-publics


Report by WHO: “This guide focuses on a specific form of citizen engagement, namely mini-publics, and their potential to be adapted to a variety of contexts. Mini-publics are forums that include a cross-section of the population selected through civic lottery to participate in evidence-informed deliberation to inform policy and action. The term refers to a diverse set of democratic innovations to engage citizens in policy-making. This guide provides an overview of how to organize mini-publics in the health sector. It is a practical companion to the 2022 Overview report, Implementing citizen engagement within evidence-informed policy-making. Both documents examine and encourage contributions that citizens can make to advance WHO’s mission to achieve universal health coverage…(More)””

i.AI Consultation Analyser


New Tool by AI.Gov.UK: “Public consultations are a critical part of the process of making laws, but analysing consultation responses is complex and very time consuming. Working with the No10 data science team (10DS), the Incubator for Artificial Intelligence (i.AI) is developing a tool to make the process of analysing public responses to government consultations faster and fairer.

The Analyser uses AI and data science techniques to automatically extract patterns and themes from the responses, and turns them into dashboards for policy makers.

The goal is for computers to do what they are best at: finding patterns and analysing large amounts of data. That means humans are free to do the work of understanding those patterns.

Screenshot showing donut chart for those who agree or disagree, and a bar chart showing popularity of prevalent themes

Government runs 700-800 consultations a year on matters of importance to the public. Some are very small, but a large consultation might attract hundreds of thousands of written responses.

A consultation attracting 30,000 responses requires a team of around 25 analysts for 3 months to analyse the data and write the report. And it’s not unheard of to get double that number

If we can apply automation in a way that is fair, effective and accountable, we could save most of that £80m…(More)”

Participatory democracy in the EU should be strengthened with a Standing Citizens’ Assembly


Article by James Mackay and Kalypso Nicolaïdis: “EU citizens have multiple participatory instruments at their disposal, from the right to petition the European Parliament (EP) to the European Citizen’s Initiative (ECI), from the European Commission’s public online consultation and Citizens’ Dialogues to the role of the European Ombudsman as an advocate for the public vis-à-vis the EU institutions.

While these mechanisms are broadly welcome they have – unfortunately – remained too timid and largely ineffective in bolstering bottom-up participation. They tend to involve experts and organised interest groups rather than ordinary citizens. They don’t encourage debates on non-experts’ policy preferences and are executed too often at the discretion of the political elites to  justify pre-existing policy decisions.

In short, they feel more like consultative mechanisms than significant democratic innovations. That’s why the EU should be bold and demonstrate its democratic leadership by institutionalising its newly-created Citizens’ Panels into a Standing Citizens’ Assembly with rotating membership chosen by lot and renewed on a regular basis…(More)”.

Digitalisation and citizen engagement: comparing participatory budgeting in Rome and Barcelona


Book chapter by Giorgia Mattei, Valentina Santolamazza and Martina Manzo: “The digitalisation of participatory budgeting (PB) is an increasing phenomenon in that digital tools could help achieve greater citizen engagement. However, comparing two similar cases – i.e. Rome and Barcelona – some differences appear during the integration of digital tools into the PB processes. The present study describes how digital tools have positively influenced PB throughout different phases, making communication more transparent, involving a wider audience, empowering people and, consequently, making citizens’ engagement more effective. Nevertheless, the research dwells on the different elements adopted to overcome the digitalisation limits and shows various approaches and results…(More)”.

Influencers: Looking beyond the consensus of the crowd.


Article by Wilfred M. McClay: “Those of us who take a loving interest in words—their etymological forebears, their many layers of meaning, their often-surprising histories—have a tendency to resist change. Not that we think playfulness should be proscribed—such pedantry would be a cure worse than any disease. It’s just that we are also drawn, like doting parents, into wanting to protect the language, and thus become suspicious of mysterious strangers, of the introduction of new words, and of new meanings for familiar ones.

When we find words being used in a novel way, our countenances tend to stiffen. What’s going on here? Is this a euphemism? Is there a hidden agenda here?

But there are times when the older language seems inadequate, and in fact may mislead us into thinking that the world has not changed. New signifiers may sometimes be necessary, in order to describe new things.

Such is unquestionably the case of the new/old word influencer. At first glance, it looks harmless and insignificant, a lazy and imprecise way of designating someone as influential. But the word’s use as a noun is the key to what is different and new about it. And much as I dislike the word, and dislike the phenomenon it describes, necessity seems to have dictated that such a word be created…(More)”.

To Design Cities Right, We Need to Focus on People


Article by Tim Keane: “Our work in the U.S. to make better neighborhoods, towns and cities is a hapless and obdurate mess. If you’ve attended a planning meeting anywhere, you have probably witnessed the miserable process in action—unrestrainedly selfish fighting about false choices and seemingly inane procedures. Rather than designing places for people, we see cities as a collection of mechanical problems with technical and legal solutions. We distract ourselves with the latest rebranded ideas about places—smart growth, resilient cities, complete streets, just cities, 15-minute cities, happy cities—rather than getting down to the actual work of designing the physical place. This lacks a fundamental vision. And it’s not succeeding.

Our flawed approach to city planning started a century ago. The first modern city plan was produced for Cincinnati in 1925 by the Technical Advisory Corporation, founded in 1913 by George Burdett Ford and E.P. Goodrich in New York City. New York adopted the country’s first comprehensive zoning ordinance in 1916, an effort Ford led. Not coincidentally, the advent of zoning, and then comprehensive planning, corresponded directly with the great migration of six million Black people from the South to Northern, Midwestern and Western cities. New city planning practices were a technical means to discriminate and exclude.

This first comprehensive plan also ushered in another type of dehumanization: city planning by formula. To justify widening downtown streets by cutting into sidewalks, engineers used a calculation that reflected the cost to operate an automobile in a congested area—including the cost of a human life, because crashes killed people. Engineers also calculated the value of a sidewalk through a formula based on how many people the elevators in adjoining buildings could deliver at peak times. In the end, Cincinnati’s planners recommended widening the streets for cars, which were becoming more common, by shrinking sidewalks. City planning became an engineering equation, and one focused on separating people and spreading the city out to the maximum extent possible…(More)”.

Six ways to democratise city planning 


Report by DemocracyNext: “To live in thriving and healthy cities, we propose six possible ways to instigate systemic changes that can democratise the governance of urban planning decisions through Citizens’ Assemblies. Depending on a city’s current starting point, at least one, if not multiple, of these options can be seen as an initial ‘way in’ to begin making systemic changes to urban planning decision making. The six ways are outlined as different entry points on the following page…(More)”.

Does information about citizen participation initiatives increase political trust?


Paper by Martin Ardanaz,  Susana Otálvaro-Ramírez, and Carlos Scartascini: “Participatory programs can reduce the informational and power asymmetries that engender mistrust. These programs, however, cannot include every citizen. Hence, it is important to evaluate if providing information about those programs could affect trust among those who do not participate. We assess the effect of an informational campaign about these programs in the context of a survey experiment conducted in the city of Buenos Aires, Argentina. Results show that providing detailed information about citizen involvement and outputs of a participatory budget initiative marginally shapes voters’ assessments of government performance and political trust. In particular, it increases voters’ perceptions about the benevolence and honesty of the government. Effects are larger for individuals with ex ante more negative views about the local government’s quality and they differ according to the respondents’ interpersonal trust and their beliefs about the ability of their communities to solve the type of collective-action problems that the program seeks to address. This article complements the literature that has examined the effects of participatory interventions on trust, and the literature that evaluates the role of information. The results in the article suggest that participatory budget programs could directly affect budget allocations and trust for those who participate, and those that are well-disseminated could also affect trust in the broader population. Because mistrustful individuals tend to shy away from demanding the government public goods that increase overall welfare, well-disseminated participatory budget programs could affect budget allocations directly and through their effect on trust…(More)”.

Developing skills for digital government


OECD “review of good practices across OECD governments”: “Digital technologies are having a profound impact on economies, labour markets and societies. They also have the potential to transform government, by enabling the implementation of more accessible and effective services. To support a shift towards digital government, investment is needed in developing the skills of civil servants. This paper reviews good practices across OECD countries to foster skills for digital government. It presents different approaches in public administration to organising training activities as well as opportunities for informal learning. It also provides insights into how relevant skills can be identified through competence frameworks, how they can be assessed, and how learning opportunities can be evaluated….(More)”