Consultation on the governance of the UN’s Scientific Advsory Panel on AI. Posted on LinkedIn.
Category: 2/ Hard Questions
Hard Questions by Sector
-
Université de l’Alberta Conférence Annuelle: Les leviers du pouvoir dans l’IA
Les leviers du pouvoir en IA
Conférencière : Tammy MacKenzie
Congrès du Campus Saint-Jean de l’Université de l’Alberta, le 25 avril 2025, Edmonton, AB (Canada). -
What We Do Not Know: GPT Use in Business and Management
This systematic review examines peer-reviewed studies on application of GPT in business management, revealing significant knowledge gaps. Despite identifying interesting research directions such as best practices, benchmarking, performance comparisons, social impacts, our analysis yields only 42 relevant studies for the 22 months since its release. There are so few studies looking at a particular sector or subfield that management researchers, business consultants, policymakers, and journalists do not yet have enough information to make well-founded statements on how GPT is being used in businesses. The primary contribution of this paper is a call to action for further research. We provide a description of current research and identify knowledge gaps on the use of GPT in business. We cover the management subfields of finance, marketing, human resources, strategy, operations, production, and analytics, excluding retail and sales. We discuss gaps in knowledge of GPT potential consequences on employment, productivity, environmental costs, oppression, and small businesses. We propose how management consultants and the media can help fill those gaps. We call for practical work on business control systems as they relate to existing and foreseeable AI-related business challenges. This work may be of interest to managers, to management researchers, and to people working on AI in society.
-
World AI: Women in AI
The Aula Fellowship were present, to discuss social and environmental concerns to do with the marketing of AI as a panacea. The event brings together technology companies, civil society, and decision-makers. We were able to connect with other non-profits and universities in this sector, and to build collaborations with several attendees and presenters.
-
‘Mind the gap’: artificial intelligence and journalism training in Southern African journalism schools
This article examines journalism schools (J-schools) responses to the Artificial Intelligence (AI) ‘disruption’. It critically provides an exploratory examination of how J-Schools in Southern Africa are responding to the AI wave in their journalism curriculums. We answer the question: How are Southern African J-Schools responding to AI in their curriculums? Using a disruptive innovation theoretical lens and through documentary review of university teaching initiatives and accredited journalism curriculums, augmented by in-depth interviews, we demonstrate that AI has opened up new horizons for journalism training in multi-dimensional ways. However, this has brought challenges, including covert forms of resistance to AI integration by some Journalism educators. Furthermore, resource constraints and the obduracy of J-schools’ curriculums also contribute to the slow introduction of AI in J-schools.
-
A labeled clinical-MRI dataset of Nigerian brains
There is currently a paucity of neuroimaging data from the African continent, limiting the diversity of data from a significant proportion of the global population. This in turn diminishes global health research and innovation. To address this issue, we present and describe the first Magnetic Resonance Imaging (MRI) dataset from individuals in the African nation of Nigeria. This dataset contains pseudonymized structural MRI (T1w, T2w, FLAIR) data of clinical quality, with 35 images from healthy control subjects, 31 images from individuals diagnosed with age-related dementia, and 22 from individuals with Parkinson’s Disease. Given the potential for Africa to contribute to the global neuroscience community, this unique MRI dataset represents both an opportunity and benchmark for future studies to share data from the African continent.
-
Community energy justice: A review of origins, convergence, and a research agenda
The transition to zero‑carbon sustainable energy systems is critical and must take an equity-oriented approach to avoid exacerbating societal injustices. We explore the concept of “community” and its potential as a viable and effective tool for studying, understanding, and fostering justice and equity in energy transitions. This paper outlines community energy justice as an area of scholarship emerging through convergence around three key concepts: community, energy transition, and justice. Using a narrative literature review approach, we unpack the origins of community energy justice research, rooted in two scholarship pillars of energy justice and community energy. We outline four driving forces and two key approaches leading to convergence between both areas of scholarship. Encompassing energy transition initiatives that incorporate both justice and community themes, we find that the overarching objective …
-
Africa’s Energy Poverty in An Artificial Intelligence (AI) World: Struggle for Sustainable Development Goal 7
Energy poverty remains a significant challenge in Sub-Saharan Africa (SSA), where approximately 600 million people lack proper access to electricity. This paper examines the region’s current state of energy poverty, highlighting its socio-economic impacts and the barriers to achieving Sustainable Development Goal 7 (SDG7), which aims for affordable, reliable, sustainable, and modern energy for all by 2030. Despite the region’s rich renewable energy potential, inadequate infrastructure, economic constraints, and governance issues continue to impede progress. This work employs a doctrinal research methodology, focusing on the critical analysis of existing legal and policy frameworks relevant to energy poverty and the integration of AI in energy management. This paper presents an overview of energy poverty in SSA, underpinned by current statistics and trends. It then examines the dual role of artificial intelligence (AI) and how it impacts this area: while AI technologies, through its data centre s, for example, significantly increase energy consumption, AI also offers innovative solutions for energy management, efficiency, and the integration of renewable energy sources. This paper critically analyzes these dynamics using Marxist and Third World Approaches to International Law (TWAIL) frameworks to understand the broader socio-economic inequalities and global power dynamics at play. Major findings indicate that current policy frameworks are inadequate in addressing the unique challenges of energy poverty and the growing role of AI in the energy sector. The paper reviews existing policy and regulatory frameworks, identifying gaps and proposing actionable recommendations for integrating AI into policies to address energy poverty. It concludes with actionable policy recommendations to achieve a just and inclusive energy transition, contributing to the broader discourse on sustainable development and technological equity.
-
Easy to read, easier to write: the politics of AI in consultancy trade research
AI systems have been rapidly implemented in all sectors, of all sizes and in every country. In this article, we conduct a bibliometric review of references in recent consultancy reports on AI use in business, policymaking, and strategic management. The uptake of these reports is high. We find three positive factors: focus on client-facing solutions, speed of production, and ease of access. We find that the evidentiary quality of reports is often unsatisfactory because of references-clubbing with other consultancy reports, references to surveys without transparency, or poor or missing references. To optimize the utility of consultancy reports for decision-makers and their pertinence for policy, we present recommendations for the quality assessment of consultancy reporting on AI’s use in organizations. We discuss how to improve general knowledge of AI use in business and policymaking, through effective collaborations between consultants and management scientists. In addition to being of interest to managers and consultants, this work may also be of interest to media, political scientists, and business-school communities.
-
Options and Motivations for International AI Benefit Sharing
Advanced AI systems could generate substantial economic and other societal benefits, but these benefits may not be widely shared by default. For a range of reasons, a number of prominent actors and institutions have called for efforts to expand access to AI’s benefits. In this report, we define the concept of international AI benefit sharing (“benefit sharing”) as efforts to support and accelerate international access to AI’s economic or broader societal benefits. Calls for benefit sharing typically invoke at least one of three motivations: 1) supporting inclusive economic growth and sustainable development, 2) fostering technological self-determination in low- and middle-income countries, and 3) advancing geopolitical objectives, including strengthening international partnerships on AI governance. Notably, as a subset of the third motive, some powerful actors – like the US government – may support benefit sharing as a tool to further their economic and national security interests. Benefit sharing could be implemented by (1) sharing AI resources (e.g., computing power or data), (2) expanding access to AI systems, or (3) transferring a portion of the financial proceeds from AI commercialisation or AI-driven economic growth. Depending on the objective that benefit sharing is intended to achieve, each of these approaches offers distinct opportunities and implementation challenges. These challenges include the potential for some benefit-sharing options to raise security concerns and increase certain global risks. Actors interested in benefit sharing may consider implementing low-risk forms of benefit sharing immediately, while launching cooperative international discussions to develop more comprehensive, mutually-beneficial initiatives.
-
UAE Abaya Fashion: From Cover to Prestige (and Social Liberalization)
This paper examines the power of the UAE abaya. Moreover, it is concerned with the exploitation of luxury in the pursuit of social status and the attainment of greater freedom within an authoritarian context. As will be argued, the abaya has transitioned from serving the state in the process of identity formation to becoming a non-state actor capable of challenging dominant strictures and providing for policy alternatives. However, while the new or revamped abaya has contributed to self-actualization and made taboo topics more visible, it is also important to note that some Emiratis or minority groups may end up being excluded from this largely luxury-driven process. For the leadership, this could create an unenviable situation, particularly when considering the potential rift between the promises outlined in the state vision and the prerequisites needed for its implementation. With this in mind, the present analysis is also intended to assist policymakers working on tolerance and social cohesion, as well as those striving to position the UAE as a major point of reference in global affairs.
-
Potential and perils of large language models as judges of unstructured textual data
Rapid advancements in large language models have unlocked remarkable capabilities when it comes to processing and summarizing unstructured text data. This has implications for the analysis of rich, open-ended datasets, such as survey responses, where LLMs hold the promise of efficiently distilling key themes and sentiments. However, as organizations increasingly turn to these powerful AI systems to make sense of textual feedback, a critical question arises, can we trust LLMs to accurately represent the perspectives contained within these text based datasets? While LLMs excel at generating human-like summaries, there is a risk that their outputs may inadvertently diverge from the true substance of the original responses. Discrepancies between the LLM-generated outputs and the actual themes present in the data could lead to flawed decision-making, with far-reaching consequences for organizations. This research investigates the effectiveness of LLM-as-judge models to evaluate the thematic alignment of summaries generated by other LLMs. We utilized an Anthropic Claude model to generate thematic summaries from open-ended survey responses, with Amazon’s Titan Express, Nova Pro, and Meta’s Llama serving as judges. This LLM-as-judge approach was compared to human evaluations using Cohen’s kappa, Spearman’s rho, and Krippendorff’s alpha, validating a scalable alternative to traditional human centric evaluation methods. Our findings reveal that while LLM-as-judge offer a scalable solution comparable to human raters, humans may still excel at detecting subtle, context-specific nuances. Our research contributes to the growing body of knowledge on AI assisted text analysis. Further, we provide recommendations for future research, emphasizing the need for careful consideration when generalizing LLM-as-judge models across various contexts and use cases.