Decision-making under uncertainty often relies on access to accurate probabilistic forecasts. In many contexts, such forecasts are scarce or difficult to obtain. Decentralised prediction markets are widely regarded as effective tools to aggregate dispersed information that decision-makers can use as forecasts to make better decisions about an uncertain future. Recently, there have been major advances in large language models that have led to claims that large language models could complement or replace market-based forecasting by synthesising information without the need for incentive-driven markets. However, there is limited empirical evidence comparing forecasts generated by large language models to market-based aggregation under real-world conditions. This thesis puts these claims to the test by examining the extent to which large language models can replicate or complement human forecasting as reflected in decentralised prediction markets. Using Polymarket as a benchmark for collective human forecasting, probability forecasts generated by large language models are compared to live market probabilities. Forecasting performance is evaluated across different market conditions, and the decision-making relevance of forecasts generated by large language models is evaluated through trading simulations. The results show that market probabilities are consistently more accurate than the forecasts generated by large language models in terms of predictive accuracy. The findings hold across all evaluated models, model combinations, prompting strategies, market stages, and liquidity levels of markets. A regression-based aggregation model that mixes market probabilities and large language model forecasts achieves predictive performance comparable to that of the market in some cases, but it fails to generalise when put to the test under realistic conditions. The findings suggest that large language models at their current stage cannot substitute prediction markets as information aggregation mechanisms. The results challenge claims that large language models can replicate the performance of prediction markets in the generation of accurate probabilistic forecasts. The results highlight the need for caution when deploying large language models in the context of high-stakes decision-making.

https://github.com/KetillHafdal/llm-vs-prediction-markets ... Read more

Earth’s orbits are becoming increasingly congested, which makes the orbital environment more fragile and harder to use safely. One of the most damaging single events is a destructive anti-satellite (ASAT) weapon test that creates long-lived debris, as these fragments threaten civil, commercial, and military space services and raise collision risks. Despite the shared danger, current international law does not explicitly ban debris-creating ASAT tests and existing accountability mechanisms remain weak. At the same time, diplomatic forums are divided between states favouring voluntary behavioural norms and states advocating a new legally binding instrument, which slows collective action and sustains political gridlock. This raises the question of how a practical and verifiable governance solution can be designed to reduce and ultimately prevent debris-creating ASAT testing under today’s geopolitical constraints, and which institutional design options the Netherlands can realistically support to improve debris prevention and accountability. This thesis treats the technical risks, legal gaps, and political barriers as a systems and institutional design challenge. Using Ostrom’s ADICO grammar, it specifies institutional rules through three core elements: agreement scope, verification, and consequences, then evaluates designs against three goals: debris prevention, accountability, and political feasibility. The design space is built from four inputs: lessons from comparative regimes (the Antarctic Treaty for transparency and a living forum, Maritime Law for precise zonal rules and dispute settlement logic, and the Nuclear Weapons Treaty for norm-building and stigmatisation), expert interviews on what is acceptable in current diplomacy and the potential role of The Hague, verification lessons from the UNIDIR handbook and the Netherlands Space Office verification study, and effect-based, technology-neutral rule reasoning focused on observable outcomes. Screening these ideas yields three contrasting scenarios: Design 1, a direct-ascent debris ban in low Earth orbit verified through national technical means and SSA with public attribution as the main consequence; Design 2, an all-kinetic debris ban across all orbits verified through a multilateral data-sharing cell with stronger collective “naming and shaming”; and Design 3, a technology-neutral ban on deliberate debris creation across all orbits verified by an independent technical body with stronger consequences such as suspension of regime benefits and privileges. The evaluation shows a clear trade-off between political feasibility and accountability, and recommends starting with Design 1 as a realistic first step that can open a pathway towards stronger, more comprehensive governance over time. The thesis also identifies concrete Dutch contributions to enable this pathway, including reframing the issue as space safety and sustainability, using middle-power diplomacy to build cross-regional coalitions (including Global South partners), promoting The Hague as an optional venue for arbitration and peaceful dispute settlement, and strengthening technical credibility through support for SSA/SST expertise and transparent verification practices. ... Read more

The primary objective is to explore the relationship between support for Universal Basic Income (latterly referred to as UBI), labor participation, and poverty reduction in the Netherlands across various financial policies. The research also seeks to examine the behavioral responses of Dutch citizens to UBI policies, considering potential benefits beyond mere financial aspects such as education, health, and volunteerism. Ultimately, the study aims to offer policy recommendations regarding UBI based on its findings.

To address these objectives, the study analyzes the impact of UBI on recipients' disposable income. Support for UBI and its influence on labor participation are investigated using conjoint analysis, which involves varying attributes such as the value of UBI per adult, UBI per child, changes in the taxation system, and modifications in allowances. Additionally, the study employs a traditional survey to gauge respondents' intentions regarding the use of time and money under two proposed UBI policies: mid-UBI and high-UBI. Furthermore, the effect of UBI on poverty is assessed by examining marginal taxation rate tables from prior research, testing five different UBI policies to determine their impact on household incomes and changes in marginal tax rates.

The findings of this study reveal mixed effects on both labor participation and support for UBI, with an average of a slight overall increase. However, there is variability in the results, with both positive and negative values present within the confidence intervals. Notably, single parents and individuals receiving welfare payments demonstrate increased labor participation across their entire confidence intervals, despite lacking support for UBI. This discrepancy can be attributed to the fact that single adult households and single parents tend to fare poorly under UBI, facing heightened poverty risks. Conversely, UBI appears to be beneficial in boosting disposable income for multi-adult households, positioning them as the primary beneficiaries of this policy.

To ensure the success of UBI, a shift in policy is imperative, moving away from incentivizing individual living— a trend that has been increasingly prominent in the Netherlands. As an initial step, UBI can be introduced modestly with a taxable monthly amount of 210 euros per month per adult, supplemented by 300 euros per month per child, and bolstered by the provision of rent allowances and free childcare. This approach aims to augment the disposable income of households earning up to 60,000 euros in gross income, without adversely affecting higher-income earners. Notably, this policy adjustment also serves to reduce marginal taxation rates from levels sometimes exceeding 100\% to a more manageable 69\%.

To progress towards higher UBI values per adult, further research is warranted into inefficient tax deductions, potential increases in income tax, considerations around land value and property tax, and the reallocation of governmental employees. Based on the findings of these investigations, UBI could be incrementally increased to 630 euros per month per adult, at which point the rent allowance may be phased out.

At the threshold of 630 euros per month, significant enhancements in quality of life are expected to ripple across the population. These improvements encompass a reduction in stress levels, augmented accessibility to healthcare services, the empowerment to opt for healthier dietary options, and an enhanced ability to save among low and middle-income earners in the Netherlands. Building on these gains, the trajectory of UBI can be extended towards a higher benchmark of 1200 euros per month per adult. This upward progression aims to unlock additional benefits, including a more optimal allocation of jobs within the Dutch populace and further mitigation of marginal taxation rates towards levels akin to standard income tax rates. ... Read more

Fast developing technology offers many opportunities for innovation and optimization in the public domain. However this sector is bound to European directives on public procurement transposed into national law and therefore bound to the mandatory procurement of, among others, their assets. This legislation provides a limiting framework in which, after awarding the contract, little room for design adjustments is perceived to be left. As a result assets, especially with long-term procurement processes, are not in accordance with current technological development. The limited amount of literature on this subject provides mitigation in very specific cases, but no overview of flexibilities in the procurement process exists, which is therefore the objective of this research. The Dutch railway sector was used as context of this research and since most process design choices are made by the procurer, the Nederlandse Spoorwegen (NS) [Dutch Railways].
First, the current process was mapped out using the functional modelling technique IDEF0 based on which a first identification of flexibilities was performed. Subsequently, a first version of a morphological chart (MC) was made. This is a product design tool, in this research used, tested and validated for institutional design, as a parallel research focus supporting the exploration of flexibilities. The MC was developed and validated on its content as well as its use as institutional design tool. Additionally, validated trends were identified in the combination of flexibility aspects, followed by the exploration of a future institutional design for the NS.
During this research four additional construction rules for the MC were found, being (1) within scope of action, (2) readability, (3) abstraction level and (4) informational value, for which a variety of practical solutions are suggested. These were adhered to in the developed MC and resulted in the validation of this tool to be useful as systematic, substantiated and concrete guidance in the discussion of process design.
Using these developed conditions for the MC enabled identification and representation of aspects enhancing flexibility, being split into three chapters being Product, Contract and Market Approach. Validated conceptual designs enhancing flexibility were created by combining different fragments of solutions into one conceptual design. Six of these were created, one being the “traditional” procurement, four were based on collaborative practices and the last was based on the product to be part of a network. These four conceptual designs based on collaboration focused on: a partnership for the separate development of innovation, international collaboration and supplier-procurer collaboration of a light and more extensive degree.
Applying this to the case of the NS shows that separate innovation and extensive collaboration are perceived to be most suitable as starting point of their future procurement design.
Altogether it can be concluded that commencing a shift from a vertical to a horizontal approach within the buyer-supplier interaction is essential, placing greater importance on formalizing the partnership rather than specifying precise outcomes. The required redesign for enhanced flexibility to intermediate design changes in public procurement processes can be obtained by using the developed MC as guidance for discussion on the process design...

... Read more

Integrating artificial intelligence (AI) into Japan's life insurance sector marks a significant move towards data-centric precision, reflecting the nation's shift towards Society 5.0. In this domain, AI is revolutionizing decision-making processes and enhancing operational efficiency, with applications ranging from fraud detection in credit card systems to predictive underwriting. While AI offers notable benefits, it also introduces risks, including privacy breaches, algorithmic biases, and inadequate human supervision. To tackle these risks, the Japanese government has initiated guidelines for societal protection, but gaps remain in the insurance industry's implementation, especially in translating social norms into the industry's context to protect stakeholders' values.

The industry needs a guide for safely designing, developing, and deploying AI systems, considering stakeholders' perspectives. This guide fills two knowledge gaps: a framework for translating high-level values into Japanese life insurance industry requirements and an initial process for converting these high-level values into organizational guidelines.

An empirical study on predictive underwriting informed the research, identifying 13 values and four informal social institutions for the AI design process. It involved eight experts who defined 54 norms, which were later refined and categorized into process and assessment norms focusing on data and AI.

The result is ten design guidelines for AI system developers, which are validated by experts, addressing the full AI lifecycle. These guidelines contribute scientifically by introducing an initial process combining design for values with system safety concepts, reporting standardization, and AI governance frameworks.
Future research should replicate this process in various contexts, reevaluate the value framework with broader stakeholder inputs, investigate the dynamics between Japanese society and AI in more detail, and delve deeper into system theoretic hazards analysis. This approach promises to strengthen the value framework and process applicability in different organizational settings.
... Read more

This paper explores the potential of smart contracts in the reinsurance industry to address escalating non-productive costs driven by market dynamics. Reinsurers, aiming to enhance stability amid increasing inflation, claim severity, and frequency, have adopted stricter underwriting criteria and raised premiums. This has led to a more detailed drafting of reinsurance contracts, reminiscent of the formalization trend that began in the 1970s. However, this formalization has inadvertently inflated administrative and dispute resolution costs, counter to the industry's core objective of efficient risk and capital allocation.

The study delves into how smart contracts can offer a solution. Smart contracts are computerized protocols that can automate contract clauses, potentially aligning better with industry goals. The examination focuses on their impact on transaction costs, involving expenses incurred by insurers and reinsurers to execute transactions.

Findings reveal that smart contracts can effectively reduce administrative costs by automating tasks, particularly in high-volume and standardized scenarios. However, their effect on dispute resolution costs is more nuanced, as the reinsurance sector still benefits from human interpretation.

In conclusion, while smart contracts hold promise for reducing transaction costs in reinsurance, the industry's unique complexities and high financial stakes may pose challenges and necessitate post-implementation adjustments. The paper recommends exploring smart contract applications in industries with smaller disputed amounts and lower trust levels than reinsurance. ... Read more

Recent technological advancements have enabled the development of increasingly impactful and complex Artificial Intelligence (AI) systems. This complexity is paired with a trade-off in terms of system opacity. The resulting lack of understanding combined with reported algorithm scandals have decreased public trust in AI systems. Meanwhile, the AI risk mitigation field is maturing. One of the proposed mechanisms to incentivize the verifiable development of trustworthy AI systems is the AI audit: the external assessment of AI systems.

The AI audit is an emerging subdomain of the Information Technology (IT) audit, a standardized practice carried out by accountants. Contrary to the IT audit, there are currently no AI-specific defined rules and regulations to adhere to. At the same time, some organizations are already seeking external assurance from accountancy firms on their AI systems. AI auditors have indicated that this has lead to challenges in their current audit approach, mainly due to a lack of structure. Therefore, this thesis proposes an AI audit workflow comprised of a general AI auditing framework combined with a structured scoping approach.

Interviews with AI auditors at one accountancy firm in the Netherlands revealed that the demand for AI audits is increasing and expected to keep growing. Clients mainly seek assurance for management of stakeholders and reputation. Furthermore, the challenges the auditors currently experience stem from having to aggregate auditing questions from a range of auditing frameworks, causing issues in their recombination and in determining question relevancy. Subsequently, design criteria for a general auditing framework as well as feedback on a proposed scoping approach were obtained.

Fourteen AI auditing frameworks were identified through a literature search. Following their typology, these could be subdivided into three source categories: academic, industry, and auditing/regulatory. Academic frameworks typically focused on specific aspects of trustworthy AI, while industry frameworks emphasized the need for public trust to drive AI progress. Frameworks developed by auditing and regulatory organizations tended to be most extensive.

Comparison to four common IT audit frameworks and standards showed that AI audit frameworks need to cover a broader range of topics than the traditional IT audit themes. This is a result of the complex socio-technological context involving multiple stakeholders in which AI systems operate. Additionally, it was shown that AI performance monitoring dashboards could cover technical parts of the audit, but that they fall short when it comes to context-dependent topics such as human oversight or societal well-being.

Following analysis of the similarities between the corporate Environmental, Social and Governance (ESG) reporting materiality assessment and the AI audit scoping problem, an ESG materiality assessment approach was translated to a scoping approach for the AI audit. In this translation, feedback from the AI auditors was incorporated. Combined with a general auditing framework, which was built through combination of the fourteen identified frameworks along the obtained design criteria, this formed the basis for the proposed AI audit workflow. The proposed workflow was demonstrated to be executable through a mock case study. Investigation from the data subject perspective for the Public Eye crowd monitoring AI system of the Municipality of Amsterdam resulted in a scoped list of auditing questions relating to privacy, transparency and fairness.

Recommendations for future AI audit workflow designs include exploring the option of incorporating subthemes in the general framework, closer co-development with AI auditors, obtaining insights from auditors at multiple accountancy firms, and automating parts of the audit. ... Read more

Alternative fuels can help the maritime shipping industry to lower its emissions. However, there are commercial risks that surround the adoption of these cleaner fuels. A book & claim based carbon insetting initiative could reduce these commercial risks and accelerate the decarbonization of the maritime shipping industry. Carbon insetting is the process of funding an emission reduction within the sector where the emissions that are being compensated originate from, through the purchase of certificates.

Book & claim facilitates the generation and trade of certificates for certain products & their properties. Combining carbon insetting with the book & claim chain of custody concept enables the generation and trade of scope 3 CO2 certificates for shippers when vessels bunker alternative fuels. Effectively enabling shippers that want their goods to be transported sustainably to pay the price premium for using alternative fuels, thereby facilitating change and sustainability in maritime shipping, without a change in logistics.

The purpose of this master’s thesis is to understand how the concept of book & claim based carbon insetting can enhance the competitive position of the port of Rotterdam for the bunkering of alternative fuels. By simultaneously understanding what potentially the most effective way is to implement this concept at the port. This is achieved through a literature review, together with a workshop conducted for Port of Rotterdam Authority (PoRA) employees.

The workshop results show that the participants have different views on how a book & claim based carbon insetting initiative should be approached. The participants’ answers covered a range, lying on two different ends of the spectrum for potential starting points.

A book & claim based carbon insetting initiative at the port of Rotterdam can potentially enhance the port’s competitive bunkering position for alternative fuels. Implementing this concept in the port would create a clear demand stream for alternative fuels, which will enable the scaling of the supply of alternative fuels within the port, improving their affordability.

Three frameworks were identified that each have their own approach to implementing book & claim based carbon insetting in the maritime shipping industry. These frameworks were evaluated based on six criteria, which were formulated from the results of the workshop. All three frameworks were evaluated for two scenarios based on the two different starting points addressed by the workshop participants.

Based on the above research results, a business strategy has been formulated for the PoRA regarding book & claim based carbon insetting. This proposed business strategy consists of general recommendations and a range of possible approaches the PoRA could take to support the concept.

Book & claim based carbon insetting for the maritime shipping industry is still in its infancy. It has the potential to lower the commercial risks surrounding the adoption of alternative fuels, accelerating the development of a market for sustainable shipping. In turn, accelerating the decarbonization of the maritime shipping industry while potentially enhancing the competitive bunkering position of the port of Rotterdam for alternative fuels if such an initiative is set up in the port.
... Read more