Lessons from the Responsible AI & Analytics for Insurance Workshop
On June 9, 2026, the Wharton School and UNSW Business School hosted the Responsible AI & Analytics for Insurance Workshop at Jon M. Huntsman Hall. The workshop brought together leaders from academia, industry, and regulation to explore how AI is reshaping the future of insurance — examining critical issues from fairness and transparency in pricing models to the growing challenges of climate risk and affordability. Catch up on presentations and lessons shared during the workshop below.
When AI Meets Insurance: Regulation, Accountability, Climate, and Affordability
Home insurance premiums are surging, insurers are retreating from high-risk markets, and AI is reshaping how risk gets priced, often in ways that are hard to see and harder to challenge. What does responsible AI look like in insurance? Who holds it accountable? And as climate losses mount, who can still afford to be covered and how to solve the affordability crisis?
- Philip Barlow, Associate Commissioner, DC Department of Insurance, Securities & Banking
- Fei Huang, Associate Professor, School of Risk and Actuarial Studies, UNSW Business School
- John Johansen, Senior Principal, Oliver Wyman
- Ben Keys, Professor of Real Estate and Professor of Finance, the Wharton School
- Kevin Werbach, Faculty Lead, Wharton Accountable AI Lab
Interpretation and Uncertainty in Machine Learning; The Fair Pricing Playbook
Discussion 1: Interpretation and Uncertainty in Machine Learning
Machine learning models can be powerful, but can we trust what they appear to tell us? This session examined popular interpretation strategies, including partial dependence plots and distillation trees, and the pitfalls that come with them — as well as deeper layers of uncertainty about what an interpretation reveals and about the model itself. The session also explored how tight bounds on subpopulation-level harm can be derived and estimated robustly, and what this means for evaluating the fairness of algorithmic decisions in practice.
- Giles Hooker, Professor, Department of Statistics and Data Science, the Wharton School
Discussion 2: The Fair Pricing Playbook
This talk presents The Fair Pricing Playbook, an open-source practical framework that translates research from actuarial science, economics, statistics, and machine learning into a concrete four-step workflow: defining a fairness criterion appropriate to the regulatory context, building a fair pricing model that meets it, measuring the welfare implications for consumers and the firm, and auditing a deployed system. Drawing on recent work in anti-discrimination insurance pricing, fairness testing, and welfare analysis, the talk examines how fairness objectives can be meaningfully and responsibly integrated into modern data-driven pricing systems.
- Fei Huang, Associate Professor, School of Risk and Actuarial Studies, UNSW Business School
Fairness for Insurance Pricing
What do we really mean by fairness and bias in insurance pricing? This session brought together industry and academic perspectives to examine the evolving landscape of fair insurance pricing across both general and life insurance, exploring the challenges of balancing predictive accuracy, regulatory compliance, transparency, and equity in modern pricing models.
- Mallika Bender, Staff Actuary, Casualty Actuarial Society
- David Schraub, Founder and CEO, David Schraub Actuarial Consultancy
AI and Insurance Market Failures
As AI becomes more central to insurer business strategies, an important question emerges: to what extent can these technologies ease or exacerbate existing market frictions — including information asymmetries, model risk and uncertainty, market power, and regulatory constraints? This session discussed the role of AI in insurance pricing, risk modeling, monitoring, and information, while examining how regulation may need to evolve as AI-driven tools become increasingly embedded across the industry.
- Pari Sastry, Assistant Professor of Finance, the Wharton School
- Adam Solomon, Assistant Professor of Finance, NYU Stern School of Business
Regulating AI in Insurance Markets
As insurers increasingly incorporate AI into nearly every facet of their operations — from underwriting and claims handling to marketing and sales — they are raising novel and difficult questions for state insurance regulators. This session provided an overview of how state insurance law and regulation are beginning to adapt to these challenges, how they are likely to evolve in the coming years, and how economic and empirical research can help shape that trajectory.
- Daniel Schwarcz, Professor, University of Minnesota Law School
The Responsible AI & Analytics for Insurance Workshop offered attendees a deeper appreciation for the opportunities and challenges AI presents for insurance markets, and highlighted the power of bringing together perspectives from academia, industry, and regulation. We’re grateful to all who participated.
We look forward to continuing these conversations in the months ahead.
This content was created with the assistance of generative AI. All AI-generated materials are reviewed and edited by the Wharton AI & Analytics Initiative to ensure accuracy, clarity, and alignment with our standards.