Related papers: Dark Speculation: Combining Qualitative and Quanti…
We explore the notion of uncertainty in the context of modern abstractive summarization models, using the tools of Bayesian Deep Learning. Our approach approximates Bayesian inference by first extending state-of-the-art summarization models…
This article introduces a conjecture that formalises a fundamental trade-off between provable correctness and broad data-mapping capacity in Artificial Intelligence (AI) systems. When an AI system is engineered for deductively watertight…
Explanation generation frameworks aim to make AI systems' decisions transparent and understandable to human users. However, generating explanations in uncertain environments characterized by incomplete information and probabilistic models…
Safety cases for frontier AI systems should provide a convincing argument, supported by evidence, that the risk of harm is within an acceptable bound. When developers author their own safety cases, confirmation bias and conflicted…
There is a substantial and ever-growing corpus of evidence and literature exploring the impacts of Artificial intelligence (AI) technologies on society, politics, and humanity as a whole. A separate, parallel body of work has explored…
Vulnerability of Frontier language models to misuse and jailbreaks has prompted the development of safety measures like filters and alignment training in an effort to ensure safety through robustness to adversarially crafted prompts. We…
Recent proposals aiming at regulating artificial intelligence (AI) and automated decision-making (ADM) suggest a particular form of risk regulation, i.e. a risk-based approach. The most salient example is the Artificial Intelligence Act…
Concept-based Models are a class of inherently explainable networks that improve upon standard Deep Neural Networks by providing a rationale behind their predictions using human-understandable `concepts'. With these models being highly…
The growing integration of large language models across professional domains transforms how experts make critical decisions in healthcare, education, and law. While significant research effort focuses on getting these systems to communicate…
Current approaches to building general-purpose AI systems tend to produce systems with both beneficial and harmful capabilities. Further progress in AI development could lead to capabilities that pose extreme risks, such as offensive cyber…
The legal profession necessitates a multidimensional approach that involves synthesizing an in-depth comprehension of a legal issue with insightful commentary based on personal experience, combined with a comprehensive understanding of…
AI agents are increasingly deployed in complex, interactive environments, yet their runtime remains a major bottleneck for training, evaluation, and real-world use. Typical agent behavior unfolds sequentially, with each action requiring an…
Although automated harmful content detection systems are frequently used to monitor online platforms, moderators and end users frequently cannot understand the logic underlying their predictions. While recent studies have focused on…
The appreciation and utilisation of risk and uncertainty can play a key role in helping to solve some of the many ethical issues that are posed by AI. Understanding the uncertainties can allow algorithms to make better decisions by…
Risk management is particularly concerned with extreme events, but analysing these events is often hindered by the scarcity of data, especially in a multivariate context. This data scarcity complicates risk management efforts. Various tools…
Outcomes of data-driven AI models cannot be assumed to be always correct. To estimate the uncertainty in these outcomes, the uncertainty wrapper framework has been proposed, which considers uncertainties related to model fit, input quality,…
When engaging in strategic decision-making, we are frequently confronted with overwhelming information and data. The situation can be further complicated when certain pieces of evidence contradict each other or become paradoxical. The…
We propose a foundational runtime actuarial layer for autonomous AI agents in which every side-effect-bearing action carries a time-consistent, counterfactual risk toll computed against a contractually fixed safe default, inside an explicit…
Large Language Models (LLMs) can produce catastrophic responses in conversational settings that pose serious risks to public safety and security. Existing evaluations often fail to fully reveal these vulnerabilities because they rely on…
AI impact assessments often stress near-term risks because human judgment degrades over longer horizons, exemplifying the Collingridge dilemma: foresight is most needed when knowledge is scarcest. To address long-term systemic risks, we…