Estimating catastrophic harms from frontier AI is hindered by deep ambiguity: many of its risks are not only unobserved but unanticipated by analysts. The central limitation of current risk analysis is the inability to populate the $\textit{catastrophic event space}$, or the set of potential large-scale harms to which probabilities might be assigned. This intractability is worsened by the $\textit{Lucretius problem}$, or the tendency to infer future risks only from past experience. We propose a process of $\textit{dark speculation}$, in which systematically generating and refining catastrophic scenarios ("qualitative" work) is coupled with estimating their likelihoods and associated damages (quantitative underwriting analysis). The idea is neither to predict the future nor to enable insurance for its own sake, but to use narrative and underwriting tools together to generate probability distributions over outcomes. We formalize this process using a simplified catastrophic Lévy stochastic framework and propose an iterative institutional design in which (1) speculation (including scenario planning) generates detailed catastrophic event narratives, (2) insurance underwriters assign probabilistic and financial parameters to these narratives, and (3) decision-makers synthesize the results into summary statistics to inform judgment. Analysis of the model reveals the value of (a) maintaining independence between speculation and underwriting, (b) analyzing multiple risk categories in parallel, and (c) generating "thick" catastrophic narrative rich in causal (counterfactual) and mitigative detail. While the approach cannot eliminate deep ambiguity, it offers a systematic approach to reason about extreme, low-probability events in frontier AI, tempering complacency and overreaction. The framework is adaptable for iterative use and can be further augmented with AI systems.

This paper develops a prudential framework for assessing the reliability of large language models (LLMs) in reinsurance. A five-pillar architecture--governance, data lineage, assurance, resilience, and regulatory alignment--translates supervisory expectations from Solvency II, SR 11-7, and guidance from EIOPA (2025), NAIC (2023), and IAIS (2024) into measurable lifecycle controls. The framework is implemented through the Reinsurance AI Reliability and Assurance Benchmark (RAIRAB), which evaluates whether governance-embedded LLMs meet prudential standards for grounding, transparency, and accountability. Across six task families, retrieval-grounded configurations achieved higher grounding accuracy (0.90), reduced hallucination and interpretive drift by roughly 40%, and nearly doubled transparency. These mechanisms lower informational frictions in risk transfer and capital allocation, showing that existing prudential doctrines already accommodate reliable AI when governance is explicit, data are traceable, and assurance is verifiable.

One of the most significant changes in recent years in the insurance sector has been the incorporation of artificial intelligence (AI) into various phases of the insurance process. From underwriting to claims processing, artificial intelligence has the potential to transform the business by increasing efficiency, lowering costs, and improving customer experience. In this article, we will look at the function of artificial intelligence in insurance and its possible impact on the sector. Underwriting is an important part of the insurance process that involves assessing potential policyholders' risks and establishing the appropriate premium. This has traditionally been a time-consuming and labor-intensive procedure, but artificial intelligence has the potential to make it faster, more efficient, and more accurate.

We all know data runs the world. The question is, can you align insurance with data? Data has always been at the heart of insurance. Although the modern commercial insurance industry may have begun with premiums calculated over a cup of coffee, it has now embraced a long list of more sophisticated analytical techniques, ranging from statistics to generalized linear models. AI in underwriting is the new shiny object in town. Let's cover if there's any merit to the hype. AI/ML can help uncover new insights from previously underutilized data, including unstructured data like text, speech, and images. It allows for using additional data during underwriting that would otherwise be unavailable or very difficult to obtain. Throughout this article, we will understand the power of AI in insurance underwriting, the benefits of underwriting automation, the future of underwriting, and everything in between.

The Underwriting Agencies Council (UAC) says annual gross written premium at Australian agencies is now around $7.5 billion, and technology-enabled firms are leading the way as the sector expands dramatically. Sydney-based GM William Legge says UAC now has more than 120 agency members, even as mergers and acquisitions created fewer, larger agencies and a build-up of "cluster groups" owning multiple specialist agency brands. As major carriers relinquish capacity in some lines, the agency market is filling gaps in coverage, Mr Legge says, and brokers have found agencies to be a trusted, reliable market that can provide responsive service, quick turn-around times, and bespoke, tailored products for hard-to-place risks. Insurance consulting firm Xceedance offers its MGA Agility Suite tailored platform to agencies, encompassing policy administration, underwriting, distribution, a broker portal and reporting functionality. Xceedance works with agencies and insurers to facilitate and support end-to-end insurance processes across claims, finance and accounting, insurance operations, catastrophe modelling, underwriting, actuarial and analytical services, policy services and data management.

To many consumers, buying life insurance can be painful. Despite insurance companies' substantial investments over the past several years in digitizing customer onboarding and policy binding, progress has been slow and incremental and, for many companies, has fallen short of expectations. Many companies have failed to meaningfully scale their efforts to modernize underwriting. The recent COVID-19 lockdowns and ongoing physical-distancing protocols reinforce the need to rethink underwriting. More than ever, insurance companies must address customer and agent frustration with the still lengthy, high-touch, manual process. With COVID-19, paramedic home visits to conduct medical exams have become highly undesirable--especially for a "pushed" product that is not immediately crucial to the customer. In this environment, risk assessment must shift toward more remote, data-driven models, while distribution must shift from in-person interactions to more online interactions.

The recent shake-out in NASDAQ-listed tech stocks spared few – not least insuretech disrupters such as Lemonade, Root and Hippo, who saw their market valuations slump an eye watering 85-90% from their peaks at one point. It wasn't difficult to see why, given aggressive and ongoing interest rate moves by the Fed and loss ratios (measuring claims incurred as a proportion of premiums sold) heading in the wrong direction. This in turn led to a substantial negative impact on earnings. Indeed, data from Capital IQ showed Root, Lemonade and Hippo collectively wracked up $1.1bn in net losses in 2021 vs. $474m two years earlier. Yet, if the travails of Lemonade, Root and Hippo offer a salutary lesson in frothy market valuations, they've also left the door open for traditional insurance providers to recapture (using third party software providers) lost market share.

"Data is the new oil." Originally coined in 2006 by the British mathematician Clive Humby, this phrase is arguably more apt today than it was then, as smartphones rival automobiles for relevance and the technology giants know more about us than we would like to admit. Just as it does for the financial services industry, the hyper-digitization of the economy presents both opportunity and potential peril for financial regulators. On the upside, reams of information are newly within their reach, filled with signals about financial system risks that regulators spend their days trying to understand. The explosion of data sheds light on global money movement, economic trends, customer onboarding decisions, quality of loan underwriting, noncompliance with regulations, financial institutions' efforts to reach the underserved, and much more. Importantly, it also contains the answers to regulators' questions about the risks of new technology itself. Digitization of finance generates novel kinds of hazards and accelerates their development. Problems can flare up between scheduled regulatory examinations and can accumulate imperceptibly beneath the surface of information reflected in traditional reports. Thanks to digitization, regulators today have a chance to gather and analyze much more data and to see much of it in something close to real time. The potential for peril arises from the concern that the regulators' current technology framework lacks the capacity to synthesize the data. The irony is that this flood of information is too much for them to handle.

Insurance is a $1.2 trillion industry in the U.S. alone, employing 2.9 million people. Historically, the insurance industry hasn’t felt the effects of digital disruption, due to a strict regulatory environment, the scale required to create a risk portfolio, and the time needed to establish trust with customers. But in a recent IBM Institute for Business Value (IBV) survey, insurance executives identified changing market forces (such as increased competition and changing customer preferences) as the top driver affecting their enterprise. The core function of the insurance industry, risk management, has gotten more complex as customer data continues to compound. Insurance companies collect data scattered across siloed business units in paper or various unstructured digital formats. In this data-rich environment, underwriting and claims management workers don’t have immediate access to the information needed for informed internal and external decision-making, leading to burnout and costly mistakes. In fact, knowledge workers spend 30% of their time finding information required to…

The term underwriting is credited to renowned Lloyd's of London. In the early days of Lloyd's, this term meant an acceptance of a part or entire risk of an event in exchange for a premium. Since then, the term has evolved with ever-changing circumstances. Insurance underwriting has always relied on data to make decisions. To be in sync with today's fast-paced and volatile world, underwriting is once again on the cusp of change.