The Role of AI in Risk Pooling to Unlock Moral Advantage
Traditional insurance architectures fail to create communities that work together to collectively uphold shared values. What if we could design a system where taking the right action was also the most supported action? By bearing risk collectively, communities can use moral advantage to enable individuals to make ethical decisions that come with high costs to themselves.
TandaPay Concept Review
In the previous post, I explained how community-based insurance was a natural solution to manage the risks associated with reporting sexual harassment. In this post, I’m going to focus on risk pooling to encourage members of a community to take a moral action which normally imposes a high cost to the individual. Risk pooling differs from insurance because it requires that the interests of the insurer and the insured are always aligned. In insurance terminology this misalignment is called moral hazard.
Moral hazard exists when there is an adversarial relationship between the insurer and the insured. TandaPay risk pools are formed when small communities of people who know one another enter into a mutual aid contract. Stricter prerequisites force a greater alignment between the community and the individual members. This alignment doesn’t merely eliminate moral hazard; it enables moral advantage which this blog post will explain in further detail.
Unfortunately risk pooling has extremely high upfront costs. To form a TandaPay community, members must write their own unique charter. Charters are complex legal documents that may be as daunting to read as a privacy policy or ToS agreement. Studies have shown that only a minority of people bother to read these before signing them. Communities that use risk pooling are not only required to read and understand charters, they are required to write and customize them. Only 18 months ago this would have been an impossible task for most communities.
The Concept of Risk Pooling
Traditional insurance policies insure risk without fostering a deeper alignment between the insurer and the insured. Risk pooling is predicated on mutual support and it requires all members to share common objectives. Using this solidarity-based approach naturally aligns the interests of every member.
Moral Advantage in Action
The “penguin effect” in the context of sexual harassment refers to the collective hesitation of victims to be the first to publicly report sexual assault or harassment, for fear of not being believed or facing disproportionate retaliation (see talk given by Jess Ladd, also first mover disadvantage). This perpetuates a status quo that favors the perpetrators. Even though there are moral reasons as well as benefits to the community for reporting misconduct when it occurs, these are frequently outweighed by the costs to those who report. Psychological, social, and systemic deterrents create an unjust dilemma where the first victim endures the greatest burdens. Overcoming this effect requires collective action to encourage reporting instead of expecting the most vulnerable to sacrifice themselves for the common good.
The penguin effect can be overcome by communities verifying sexual harassment grievance claims themselves without forcing individuals to report claims to authorities outside of the community. The current reporting process isolates victims and often retraumatizes them. It can frequently put them at risk of retaliation and subject them to harsh ridicule. These downsides to reporting cannot be completely mitigated but community validation is likely to increase the rate at which incidents are reported. Wrongdoing uncertainty is also likely to decrease; this term refers to a situation where a perpetrator acts in a morally ambiguous manner (see Fed Up With Sexual Harassment: The Serial Harasser’s Playbook and Sexual harassment and assault in Astronomy and Physics).
Moral advantage exists when communities use risk pooling to incentivize risk taking behavior in situations where the costs of undertaking a moral act is high, like the reporting of harassment. TandaPay and similar protocols could potentially offer tremendous benefits to communities and their members. But it requires communities to clear some major hurdles first.
Writing Charters and Reading Charters to Evaluate a Claim’s Validity is Hard
Crafting a community charter is a complex endeavor akin to drafting a legally binding document. A charter goes beyond the generality of a terms of service. Charters encapsulate the ethos and operational rules for evaluating sexual harassment claims and they also articulate the heart and soul of how to care for victims. They are simultaneously statutory and sentimental. How can one document direct members to diligently examine claims for validity while also coordinating members to care for victims?
These documents are crucial for outlining a claims eligibility and for managing a community’s affairs. Every member must use these clearly written standards to scrutinize a claim’s validity while being bound to exercise compassion towards victims. The charter’s requirement that every member use them to independently verify a claim seems unreasonable. Even if the charter’s guidelines were sufficiently clear it is highly unlikely that more than 20% of members would ever want to perform claim evaluations.
Writing such a charter is no small feat, particularly for a community without extensive legal expertise. It involves translating often intricate policies — like an organization’s sexual harassment policy or a university’s Title IX policy — into actionable, community-guided procedures. The process requires a deep understanding of legal nuances and the ability to foresee and articulate responses to a myriad of potential scenarios.
LLMs Make Writing Charters and Evaluating Claims Easier
The advent of Large Language Models (LLMs) is revolutionizing the way communities approach the daunting task of charter creation. These advanced AI systems, trained on vast corpuses of legal and policy texts, are essential for transforming the raw material of employer policies and university standards into the woven fabric of a community’s charter. Community leaders can provide the LLM with specific directives. The AI then uses these to generate charter language that is both precise in its legal adherence and reflexive of the community’s values.
LLMs not only aid in the creation of these foundational documents but also simplifying the claim evaluation process. By interacting with the charter through guided dialogue with the AI, policyholders are empowered to make informed decisions on claim validity. The LLM, serving as an interactive guide, ensures that the charter’s stipulations are understood and that each policyholder’s decision on a claim is grounded in the community’s agreed-upon standards.
The Role of Large Language Models in Community Coordination
With the charter and pledge in place, the LLM’s role extends into the realm of community coordination. It becomes a mediator between policyholder and policy, providing a conversational interface that guides members through the claim evaluation process. This AI-driven dialogue ensures that policyholders’ evaluations are thorough, unbiased, and aligned with the charter’s requirements. The LLM asks pertinent questions, prompts the review of evidence, and ensures that the policyholder’s decision-making process is both rigorous and autonomous. This innovative use of technology transforms charters into sovereign entities which can demand that every policyholder uphold their pledge to serve the charter’s mandate faithfully. With the help of the LLM a policyholder can understand and meet the charter’s requirements with confidence and clarity.
The integration of LLMs into the TandaPay framework marks a pivotal step forward in community-based risk pooling. By simplifying the complexities of charter writing and claim evaluation, LLMs are unlocking the potential of ethical, transparent, and self-governed communities equipped to tackle the challenges of moral action in a supportive and structured environment.
Every Member is Capable of Reaching an Individual Evaluation
Each member participates in an independent evaluation guided completely by the community’s charter — a living document to which they have sworn unwavering fidelity. A Large Language Model (LLM) can transform a charter document into a charter entity which assists policyholders with the task of claim evaluation. Members ask clarifying questions of their AI assistant to ascertain the validity of a claim. The charter, augmented by the LLM, queries the policyholder on their review of the evidence and makes demands upon the policyholder to ascertain that the claimant has met specific requirements. Through this interactive process, the policyholder independently determines if the claim meets the standards of validity outlined in the charter. They do this without relinquishing personal responsibility to another member of their group simply because evaluating claims is technically complex. This system does not eradicate groupthink but empowers members to uphold their pledge to personally assess each claim.
Barriers and Breakthroughs
Historically, the concept of risk pooling at a community level faced significant barriers, both financial and regulatory. Communities looking to engage in risk pooling were often hindered by the lack of tools and frameworks to manage collective funds effectively. However, with advancements in blockchain technology and the development of platforms like TandaPay, these barriers are being dismantled. This technological evolution, coupled with the expertise provided by LLMs in drafting and interpreting community charters, is ushering in a new era of community-driven risk management.
Conclusion
The emergence of TandaPay represents a significant leap forward in enabling communities to self-govern and manage risks collectively. This model is built on the foundation of moral advantage, aligning community interests, and empowering individuals through the shared responsibility of risk pooling. The incorporation of LLMs into this framework enhances the capability of community members to engage with community charters to evaluate claims in a way that was not previously possible. As we move away from traditional insurance models and embrace the concept of risk pooling, we pave the way for a more equitable, supportive, and ethical approach to managing the risks associated with moral actions.
The integration of LLMs into peer-to-peer insurance architecture is a significant step in the evolution of community-based risk pooling. LLMs greatly simplify the complexities of charter writing and claim evaluation. This unlocks the potential for ethical, transparent, and self-governed communities ready to embrace moral action with collective support.
